Limin Wang

Zhan Tong, Yibing Song, Jue Wang et al.

Pre-training video transformers on extra large-scale datasets is generally required to achieve premier performance on relatively small datasets. In this paper, we show that video masked autoencoders (VideoMAE) are data-efficient learners for self-supervised video pre-training (SSVP). We are inspired by the recent ImageMAE and propose customized video tube masking with an extremely high ratio. This simple design makes video reconstruction a more challenging self-supervision task, thus encouraging extracting more effective video representations during this pre-training process. We obtain three important findings on SSVP: (1) An extremely high proportion of masking ratio (i.e., 90% to 95%) still yields favorable performance of VideoMAE. The temporally redundant video content enables a higher masking ratio than that of images. (2) VideoMAE achieves impressive results on very small datasets (i.e., around 3k-4k videos) without using any extra data. (3) VideoMAE shows that data quality is more important than data quantity for SSVP. Domain shift between pre-training and target datasets is an important issue. Notably, our VideoMAE with the vanilla ViT can achieve 87.4% on Kinetics-400, 75.4% on Something-Something V2, 91.3% on UCF101, and 62.6% on HMDB51, without using any extra data. Code is available at https://github.com/MCG-NJU/VideoMAE.

Yutao Cui, Cheng Jiang, Gangshan Wu et al.

Visual object tracking often employs a multi-stage pipeline of feature extraction, target information integration, and bounding box estimation. To simplify this pipeline and unify the process of feature extraction and target information integration, in this paper, we present a compact tracking framework, termed as MixFormer, built upon transformers. Our core design is to utilize the flexibility of attention operations, and propose a Mixed Attention Module (MAM) for simultaneous feature extraction and target information integration. This synchronous modeling scheme allows to extract target-specific discriminative features and perform extensive communication between target and search area. Based on MAM, we build our MixFormer trackers simply by stacking multiple MAMs and placing a localization head on top. Specifically, we instantiate two types of MixFormer trackers, a hierarchical tracker MixCvT, and a non-hierarchical tracker MixViT. For these two trackers, we investigate a series of pre-training methods and uncover the different behaviors between supervised pre-training and self-supervised pre-training in our MixFormer trackers. We also extend the masked pre-training to our MixFormer trackers and design the competitive TrackMAE pre-training technique. Finally, to handle multiple target templates during online tracking, we devise an asymmetric attention scheme in MAM to reduce computational cost, and propose an effective score prediction module to select high-quality templates. Our MixFormer trackers set a new state-of-the-art performance on seven tracking benchmarks, including LaSOT, TrackingNet, VOT2020, GOT-10k, OTB100 and UAV123. In particular, our MixViT-L achieves AUC score of 73.3% on LaSOT, 86.1% on TrackingNet, EAO of 0.584 on VOT2020, and AO of 75.7% on GOT-10k. Code and trained models are publicly available at https://github.com/MCG-NJU/MixFormer.

Yutao Cui, Cheng Jiang, Limin Wang et al.

Tracking often uses a multi-stage pipeline of feature extraction, target information integration, and bounding box estimation. To simplify this pipeline and unify the process of feature extraction and target information integration, we present a compact tracking framework, termed as MixFormer, built upon transformers. Our core design is to utilize the flexibility of attention operations, and propose a Mixed Attention Module (MAM) for simultaneous feature extraction and target information integration. This synchronous modeling scheme allows to extract target-specific discriminative features and perform extensive communication between target and search area. Based on MAM, we build our MixFormer tracking framework simply by stacking multiple MAMs with progressive patch embedding and placing a localization head on top. In addition, to handle multiple target templates during online tracking, we devise an asymmetric attention scheme in MAM to reduce computational cost, and propose an effective score prediction module to select high-quality templates. Our MixFormer sets a new state-of-the-art performance on five tracking benchmarks, including LaSOT, TrackingNet, VOT2020, GOT-10k, and UAV123. In particular, our MixFormer-L achieves NP score of 79.9% on LaSOT, 88.9% on TrackingNet and EAO of 0.555 on VOT2020. We also perform in-depth ablation studies to demonstrate the effectiveness of simultaneous feature extraction and information integration. Code and trained models are publicly available at https://github.com/MCG-NJU/MixFormer.

Limin Wang, Bingkun Huang, Zhiyu Zhao et al.

Scale is the primary factor for building a powerful foundation model that could well generalize to a variety of downstream tasks. However, it is still challenging to train video foundation models with billions of parameters. This paper shows that video masked autoencoder (VideoMAE) is a scalable and general self-supervised pre-trainer for building video foundation models. We scale the VideoMAE in both model and data with a core design. Specifically, we present a dual masking strategy for efficient pre-training, with an encoder operating on a subset of video tokens and a decoder processing another subset of video tokens. Although VideoMAE is very efficient due to high masking ratio in encoder, masking decoder can still further reduce the overall computational cost. This enables the efficient pre-training of billion-level models in video. We also use a progressive training paradigm that involves an initial pre-training on a diverse multi-sourced unlabeled dataset, followed by a post-pre-training on a mixed labeled dataset. Finally, we successfully train a video ViT model with a billion parameters, which achieves a new state-of-the-art performance on the datasets of Kinetics (90.0% on K400 and 89.9% on K600) and Something-Something (68.7% on V1 and 77.0% on V2). In addition, we extensively verify the pre-trained video ViT models on a variety of downstream tasks, demonstrating its effectiveness as a general video representation learner. The code and model is available at \url{https://github.com/OpenGVLab/VideoMAEv2}.

Zhuoyan Luo, Fengyuan Shi, Yixiao Ge et al. · tencent-ai

The Open-MAGVIT2 project produces an open-source replication of Google's MAGVIT-v2 tokenizer, a tokenizer with a super-large codebook (i.e., $2^{18}$ codes), and achieves the state-of-the-art reconstruction performance on ImageNet and UCF benchmarks. We also provide a tokenizer pre-trained on large-scale data, significantly outperforming Cosmos on zero-shot benchmarks (1.93 vs. 0.78 rFID on ImageNet original resolution). Furthermore, we explore its application in plain auto-regressive models to validate scalability properties, producing a family of auto-regressive image generation models ranging from 300M to 1.5B. To assist auto-regressive models in predicting with a super-large vocabulary, we factorize it into two sub-vocabulary of different sizes by asymmetric token factorization, and further introduce ``next sub-token prediction'' to enhance sub-token interaction for better generation quality. We release all models and codes to foster innovation and creativity in the field of auto-regressive visual generation.

Yi Wang, Kunchang Li, Yizhuo Li et al.

The foundation models have recently shown excellent performance on a variety of downstream tasks in computer vision. However, most existing vision foundation models simply focus on image-level pretraining and adpation, which are limited for dynamic and complex video-level understanding tasks. To fill the gap, we present general video foundation models, InternVideo, by taking advantage of both generative and discriminative self-supervised video learning. Specifically, InternVideo efficiently explores masked video modeling and video-language contrastive learning as the pretraining objectives, and selectively coordinates video representations of these two complementary frameworks in a learnable manner to boost various video applications. Without bells and whistles, InternVideo achieves state-of-the-art performance on 39 video datasets from extensive tasks including video action recognition/detection, video-language alignment, and open-world video applications. Especially, our methods can obtain 91.1% and 77.2% top-1 accuracy on the challenging Kinetics-400 and Something-Something V2 benchmarks, respectively. All of these results effectively show the generality of our InternVideo for video understanding. The code will be released at https://github.com/OpenGVLab/InternVideo .

Kunchang Li, Yali Wang, Yizhuo Li et al.

Video Foundation Models (VFMs) have received limited exploration due to high computational costs and data scarcity. Previous VFMs rely on Image Foundation Models (IFMs), which face challenges in transferring to the video domain. Although VideoMAE has trained a robust ViT from limited data, its low-level reconstruction poses convergence difficulties and conflicts with high-level cross-modal alignment. This paper proposes a training-efficient method for temporal-sensitive VFMs that integrates the benefits of existing methods. To increase data efficiency, we mask out most of the low-semantics video tokens, but selectively align the unmasked tokens with IFM, which serves as the UnMasked Teacher (UMT). By providing semantic guidance, our method enables faster convergence and multimodal friendliness. With a progressive pre-training framework, our model can handle various tasks including scene-related, temporal-related, and complex video-language understanding. Using only public sources for pre-training in 6 days on 32 A100 GPUs, our scratch-built ViT-L/16 achieves state-of-the-art performances on various video tasks. The code and models will be released at https://github.com/OpenGVLab/unmasked_teacher.

Kunchang Li, Yali Wang, Yinan He et al.

With the rapid development of Multi-modal Large Language Models (MLLMs), a number of diagnostic benchmarks have recently emerged to evaluate the comprehension capabilities of these models. However, most benchmarks predominantly assess spatial understanding in the static image tasks, while overlooking temporal understanding in the dynamic video tasks. To alleviate this issue, we introduce a comprehensive Multi-modal Video understanding Benchmark, namely MVBench, which covers 20 challenging video tasks that cannot be effectively solved with a single frame. Specifically, we first introduce a novel static-to-dynamic method to define these temporal-related tasks. By transforming various static tasks into dynamic ones, we enable the systematic generation of video tasks that require a broad spectrum of temporal skills, ranging from perception to cognition. Then, guided by the task definition, we automatically convert public video annotations into multiple-choice QA to evaluate each task. On one hand, such a distinct paradigm allows us to build MVBench efficiently, without much manual intervention. On the other hand, it guarantees evaluation fairness with ground-truth video annotations, avoiding the biased scoring of LLMs. Moreover, we further develop a robust video MLLM baseline, i.e., VideoChat2, by progressive multi-modal training with diverse instruction-tuning data. The extensive results on our MVBench reveal that, the existing MLLMs are far from satisfactory in temporal understanding, while our VideoChat2 largely surpasses these leading models by over 15% on MVBench. All models and data are available at https://github.com/OpenGVLab/Ask-Anything.

Kunchang Li, Yali Wang, Yinan He et al.

Learning discriminative spatiotemporal representation is the key problem of video understanding. Recently, Vision Transformers (ViTs) have shown their power in learning long-term video dependency with self-attention. Unfortunately, they exhibit limitations in tackling local video redundancy, due to the blind global comparison among tokens. UniFormer has successfully alleviated this issue, by unifying convolution and self-attention as a relation aggregator in the transformer format. However, this model has to require a tiresome and complicated image-pretraining phrase, before being finetuned on videos. This blocks its wide usage in practice. On the contrary, open-sourced ViTs are readily available and well-pretrained with rich image supervision. Based on these observations, we propose a generic paradigm to build a powerful family of video networks, by arming the pretrained ViTs with efficient UniFormer designs. We call this family UniFormerV2, since it inherits the concise style of the UniFormer block. But it contains brand-new local and global relation aggregators, which allow for preferable accuracy-computation balance by seamlessly integrating advantages from both ViTs and UniFormer. Without any bells and whistles, our UniFormerV2 gets the state-of-the-art recognition performance on 8 popular video benchmarks, including scene-related Kinetics-400/600/700 and Moments in Time, temporal-related Something-Something V1/V2, untrimmed ActivityNet and HACS. In particular, it is the first model to achieve 90% top-1 accuracy on Kinetics-400, to our best knowledge. Code will be available at https://github.com/OpenGVLab/UniFormerV2.

Yating Tian, Hongwen Zhang, Yebin Liu et al.

Estimating human pose and shape from monocular images is a long-standing problem in computer vision. Since the release of statistical body models, 3D human mesh recovery has been drawing broader attention. With the same goal of obtaining well-aligned and physically plausible mesh results, two paradigms have been developed to overcome challenges in the 2D-to-3D lifting process: i) an optimization-based paradigm, where different data terms and regularization terms are exploited as optimization objectives; and ii) a regression-based paradigm, where deep learning techniques are embraced to solve the problem in an end-to-end fashion. Meanwhile, continuous efforts are devoted to improving the quality of 3D mesh labels for a wide range of datasets. Though remarkable progress has been achieved in the past decade, the task is still challenging due to flexible body motions, diverse appearances, complex environments, and insufficient in-the-wild annotations. To the best of our knowledge, this is the first survey that focuses on the task of monocular 3D human mesh recovery. We start with the introduction of body models and then elaborate recovery frameworks and training objectives by providing in-depth analyses of their strengths and weaknesses. We also summarize datasets, evaluation metrics, and benchmark results. Open issues and future directions are discussed in the end, hoping to motivate researchers and facilitate their research in this area. A regularly updated project page can be found at https://github.com/tinatiansjz/hmr-survey.

Guozhen Zhang, Yuhan Zhu, Haonan Wang et al.

Effectively extracting inter-frame motion and appearance information is important for video frame interpolation (VFI). Previous works either extract both types of information in a mixed way or elaborate separate modules for each type of information, which lead to representation ambiguity and low efficiency. In this paper, we propose a novel module to explicitly extract motion and appearance information via a unifying operation. Specifically, we rethink the information process in inter-frame attention and reuse its attention map for both appearance feature enhancement and motion information extraction. Furthermore, for efficient VFI, our proposed module could be seamlessly integrated into a hybrid CNN and Transformer architecture. This hybrid pipeline can alleviate the computational complexity of inter-frame attention as well as preserve detailed low-level structure information. Experimental results demonstrate that, for both fixed- and arbitrary-timestep interpolation, our method achieves state-of-the-art performance on various datasets. Meanwhile, our approach enjoys a lighter computation overhead over models with close performance. The source code and models are available at https://github.com/MCG-NJU/EMA-VFI.

Ruopeng Gao, Limin Wang

As a video task, Multiple Object Tracking (MOT) is expected to capture temporal information of targets effectively. Unfortunately, most existing methods only explicitly exploit the object features between adjacent frames, while lacking the capacity to model long-term temporal information. In this paper, we propose MeMOTR, a long-term memory-augmented Transformer for multi-object tracking. Our method is able to make the same object's track embedding more stable and distinguishable by leveraging long-term memory injection with a customized memory-attention layer. This significantly improves the target association ability of our model. Experimental results on DanceTrack show that MeMOTR impressively surpasses the state-of-the-art method by 7.9% and 13.0% on HOTA and AssA metrics, respectively. Furthermore, our model also outperforms other Transformer-based methods on association performance on MOT17 and generalizes well on BDD100K. Code is available at https://github.com/MCG-NJU/MeMOTR.

Ziqi Huang, Yinan He, Jiashuo Yu et al.

Video generation has witnessed significant advancements, yet evaluating these models remains a challenge. A comprehensive evaluation benchmark for video generation is indispensable for two reasons: 1) Existing metrics do not fully align with human perceptions; 2) An ideal evaluation system should provide insights to inform future developments of video generation. To this end, we present VBench, a comprehensive benchmark suite that dissects "video generation quality" into specific, hierarchical, and disentangled dimensions, each with tailored prompts and evaluation methods. VBench has three appealing properties: 1) Comprehensive Dimensions: VBench comprises 16 dimensions in video generation (e.g., subject identity inconsistency, motion smoothness, temporal flickering, and spatial relationship, etc). The evaluation metrics with fine-grained levels reveal individual models' strengths and weaknesses. 2) Human Alignment: We also provide a dataset of human preference annotations to validate our benchmarks' alignment with human perception, for each evaluation dimension respectively. 3) Valuable Insights: We look into current models' ability across various evaluation dimensions, and various content types. We also investigate the gaps between video and image generation models. We will open-source VBench, including all prompts, evaluation methods, generated videos, and human preference annotations, and also include more video generation models in VBench to drive forward the field of video generation.

Liang Zhao, Limin Wang

Detectors trained with massive labeled data often exhibit dramatic performance degradation in some particular scenarios with data distribution gap. To alleviate this problem of domain shift, conventional wisdom typically concentrates solely on reducing the discrepancy between the source and target domains via attached domain classifiers, yet ignoring the difficulty of such transferable features in coping with both classification and localization subtasks in object detection. To address this issue, in this paper, we propose Task-specific Inconsistency Alignment (TIA), by developing a new alignment mechanism in separate task spaces, improving the performance of the detector on both subtasks. Specifically, we add a set of auxiliary predictors for both classification and localization branches, and exploit their behavioral inconsistencies as finer-grained domain-specific measures. Then, we devise task-specific losses to align such cross-domain disagreement of both subtasks. By optimizing them individually, we are able to well approximate the category- and boundary-wise discrepancies in each task space, and therefore narrow them in a decoupled manner. TIA demonstrates superior results on various scenarios to the previous state-of-the-art methods. It is also observed that both the classification and localization capabilities of the detector are sufficiently strengthened, further demonstrating the effectiveness of our TIA method. Code and trained models are publicly available at https://github.com/MCG-NJU/TIA.

Min Yang, Guo Chen, Yin-Dong Zheng et al.

Temporal action detection (TAD) is extensively studied in the video understanding community by generally following the object detection pipeline in images. However, complex designs are not uncommon in TAD, such as two-stream feature extraction, multi-stage training, complex temporal modeling, and global context fusion. In this paper, we do not aim to introduce any novel technique for TAD. Instead, we study a simple, straightforward, yet must-known baseline given the current status of complex design and low detection efficiency in TAD. In our simple baseline (termed BasicTAD), we decompose the TAD pipeline into several essential components: data sampling, backbone design, neck construction, and detection head. We extensively investigate the existing techniques in each component for this baseline, and more importantly, perform end-to-end training over the entire pipeline thanks to the simplicity of design. As a result, this simple BasicTAD yields an astounding and real-time RGB-Only baseline very close to the state-of-the-art methods with two-stream inputs. In addition, we further improve the BasicTAD by preserving more temporal and spatial information in network representation (termed as PlusTAD). Empirical results demonstrate that our PlusTAD is very efficient and significantly outperforms the previous methods on the datasets of THUMOS14 and FineAction. Meanwhile, we also perform in-depth visualization and error analysis on our proposed method and try to provide more insights on the TAD problem. Our approach can serve as a strong baseline for future TAD research. The code and model will be released at https://github.com/MCG-NJU/BasicTAD.

Keqiang Sun, Junting Pan, Yuying Ge et al. · pku

While recent advancements in vision-language models have had a transformative impact on multi-modal comprehension, the extent to which these models possess the ability to comprehend generated images remains uncertain. Synthetic images, in comparison to real data, encompass a higher level of diversity in terms of both content and style, thereby presenting significant challenges for the models to fully grasp. In light of this challenge, we introduce a comprehensive dataset, referred to as JourneyDB, that caters to the domain of generative images within the context of multi-modal visual understanding. Our meticulously curated dataset comprises 4 million distinct and high-quality generated images, each paired with the corresponding text prompts that were employed in their creation. Furthermore, we additionally introduce an external subset with results of another 22 text-to-image generative models, which makes JourneyDB a comprehensive benchmark for evaluating the comprehension of generated images. On our dataset, we have devised four benchmarks to assess the performance of generated image comprehension in relation to both content and style interpretation. These benchmarks encompass prompt inversion, style retrieval, image captioning, and visual question answering. Lastly, we evaluate the performance of state-of-the-art multi-modal models when applied to the JourneyDB dataset, providing a comprehensive analysis of their strengths and limitations in comprehending generated content. We anticipate that the proposed dataset and benchmarks will facilitate further research in the field of generative content understanding. The dataset is publicly available at https://journeydb.github.io.

Lei Chen, Zhan Tong, Yibing Song et al.

Streaming video clips with large-scale video tokens impede vision transformers (ViTs) for efficient recognition, especially in video action detection where sufficient spatiotemporal representations are required for precise actor identification. In this work, we propose an end-to-end framework for efficient video action detection (EVAD) based on vanilla ViTs. Our EVAD consists of two specialized designs for video action detection. First, we propose a spatiotemporal token dropout from a keyframe-centric perspective. In a video clip, we maintain all tokens from its keyframe, preserve tokens relevant to actor motions from other frames, and drop out the remaining tokens in this clip. Second, we refine scene context by leveraging remaining tokens for better recognizing actor identities. The region of interest (RoI) in our action detector is expanded into temporal domain. The captured spatiotemporal actor identity representations are refined via scene context in a decoder with the attention mechanism. These two designs make our EVAD efficient while maintaining accuracy, which is validated on three benchmark datasets (i.e., AVA, UCF101-24, JHMDB). Compared to the vanilla ViT backbone, our EVAD reduces the overall GFLOPs by 43% and improves real-time inference speed by 40% with no performance degradation. Moreover, even at similar computational costs, our EVAD can improve the performance by 1.1 mAP with higher resolution inputs. Code is available at https://github.com/MCG-NJU/EVAD.

Guo Chen, Sen Xing, Zhe Chen et al.

In this report, we present our champion solutions to five tracks at Ego4D challenge. We leverage our developed InternVideo, a video foundation model, for five Ego4D tasks, including Moment Queries, Natural Language Queries, Future Hand Prediction, State Change Object Detection, and Short-term Object Interaction Anticipation. InternVideo-Ego4D is an effective paradigm to adapt the strong foundation model to the downstream ego-centric video understanding tasks with simple head designs. In these five tasks, the performance of InternVideo-Ego4D comprehensively surpasses the baseline methods and the champions of CVPR2022, demonstrating the powerful representation ability of InternVideo as a video foundation model. Our code will be released at https://github.com/OpenGVLab/ego4d-eccv2022-solutions

Ziteng Gao, Zhan Tong, Limin Wang et al.

Human visual recognition is a sparse process, where only a few salient visual cues are attended to rather than traversing every detail uniformly. However, most current vision networks follow a dense paradigm, processing every single visual unit (e.g,, pixel or patch) in a uniform manner. In this paper, we challenge this dense paradigm and present a new method, coined SparseFormer, to imitate human's sparse visual recognition in an end-to-end manner. SparseFormer learns to represent images using a highly limited number of tokens (down to 49) in the latent space with sparse feature sampling procedure instead of processing dense units in the original pixel space. Therefore, SparseFormer circumvents most of dense operations on the image space and has much lower computational costs. Experiments on the ImageNet classification benchmark dataset show that SparseFormer achieves performance on par with canonical or well-established models while offering better accuracy-throughput tradeoff. Moreover, the design of our network can be easily extended to the video classification with promising performance at lower computational costs. We hope that our work can provide an alternative way for visual modeling and inspire further research on sparse neural architectures. The code will be publicly available at https://github.com/showlab/sparseformer

Tao Lu, Xiang Ding, Haisong Liu et al.

Extending the success of 2D Large Kernel to 3D perception is challenging due to: 1. the cubically-increasing overhead in processing 3D data; 2. the optimization difficulties from data scarcity and sparsity. Previous work has taken the first step to scale up the kernel size from 3x3x3 to 7x7x7 by introducing block-shared weights. However, to reduce the feature variations within a block, it only employs modest block size and fails to achieve larger kernels like the 21x21x21. To address this issue, we propose a new method, called LinK, to achieve a wider-range perception receptive field in a convolution-like manner with two core designs. The first is to replace the static kernel matrix with a linear kernel generator, which adaptively provides weights only for non-empty voxels. The second is to reuse the pre-computed aggregation results in the overlapped blocks to reduce computation complexity. The proposed method successfully enables each voxel to perceive context within a range of 21x21x21. Extensive experiments on two basic perception tasks, 3D object detection and 3D semantic segmentation, demonstrate the effectiveness of our method. Notably, we rank 1st on the public leaderboard of the 3D detection benchmark of nuScenes (LiDAR track), by simply incorporating a LinK-based backbone into the basic detector, CenterPoint. We also boost the strong segmentation baseline's mIoU with 2.7% in the SemanticKITTI test set. Code is available at https://github.com/MCG-NJU/LinK.

Jiange Yang, Sheng Guo, Gangshan Wu et al.

Current RGB-D scene recognition approaches often train two standalone backbones for RGB and depth modalities with the same Places or ImageNet pre-training. However, the pre-trained depth network is still biased by RGB-based models which may result in a suboptimal solution. In this paper, we present a single-model self-supervised hybrid pre-training framework for RGB and depth modalities, termed as CoMAE. Our CoMAE presents a curriculum learning strategy to unify the two popular self-supervised representation learning algorithms: contrastive learning and masked image modeling. Specifically, we first build a patch-level alignment task to pre-train a single encoder shared by two modalities via cross-modal contrastive learning. Then, the pre-trained contrastive encoder is passed to a multi-modal masked autoencoder to capture the finer context features from a generative perspective. In addition, our single-model design without requirement of fusion module is very flexible and robust to generalize to unimodal scenario in both training and testing phases. Extensive experiments on SUN RGB-D and NYUDv2 datasets demonstrate the effectiveness of our CoMAE for RGB and depth representation learning. In addition, our experiment results reveal that CoMAE is a data-efficient representation learner. Although we only use the small-scale and unlabeled training set for pre-training, our CoMAE pre-trained models are still competitive to the state-of-the-art methods with extra large-scale and supervised RGB dataset pre-training. Code will be released at https://github.com/MCG-NJU/CoMAE.

Haoyue Cheng, Zhaoyang Liu, Hang Zhou et al.

This paper focuses on the weakly-supervised audio-visual video parsing task, which aims to recognize all events belonging to each modality and localize their temporal boundaries. This task is challenging because only overall labels indicating the video events are provided for training. However, an event might be labeled but not appear in one of the modalities, which results in a modality-specific noisy label problem. In this work, we propose a training strategy to identify and remove modality-specific noisy labels dynamically. It is motivated by two key observations: 1) networks tend to learn clean samples first; and 2) a labeled event would appear in at least one modality. Specifically, we sort the losses of all instances within a mini-batch individually in each modality, and then select noisy samples according to the relationships between intra-modal and inter-modal losses. Besides, we also propose a simple but valid noise ratio estimation method by calculating the proportion of instances whose confidence is below a preset threshold. Our method makes large improvements over the previous state of the arts (e.g. from 60.0\% to 63.8\% in segment-level visual metric), which demonstrates the effectiveness of our approach. Code and trained models are publicly available at \url{https://github.com/MCG-NJU/JoMoLD}.

Haisong Liu, Tao Lu, Yihui Xu et al.

In this paper, we study the problem of jointly estimating the optical flow and scene flow from synchronized 2D and 3D data. Previous methods either employ a complex pipeline that splits the joint task into independent stages, or fuse 2D and 3D information in an ``early-fusion'' or ``late-fusion'' manner. Such one-size-fits-all approaches suffer from a dilemma of failing to fully utilize the characteristic of each modality or to maximize the inter-modality complementarity. To address the problem, we propose a novel end-to-end framework, which consists of 2D and 3D branches with multiple bidirectional fusion connections between them in specific layers. Different from previous work, we apply a point-based 3D branch to extract the LiDAR features, as it preserves the geometric structure of point clouds. To fuse dense image features and sparse point features, we propose a learnable operator named bidirectional camera-LiDAR fusion module (Bi-CLFM). We instantiate two types of the bidirectional fusion pipeline, one based on the pyramidal coarse-to-fine architecture (dubbed CamLiPWC), and the other one based on the recurrent all-pairs field transforms (dubbed CamLiRAFT). On FlyingThings3D, both CamLiPWC and CamLiRAFT surpass all existing methods and achieve up to a 47.9\% reduction in 3D end-point-error from the best published result. Our best-performing model, CamLiRAFT, achieves an error of 4.26\% on the KITTI Scene Flow benchmark, ranking 1st among all submissions with much fewer parameters. Besides, our methods have strong generalization performance and the ability to handle non-rigid motion. Code is available at https://github.com/MCG-NJU/CamLiFlow.

Sheng Guo, Zihua Xiong, Yujie Zhong et al.

In this paper, we present a new cross-architecture contrastive learning (CACL) framework for self-supervised video representation learning. CACL consists of a 3D CNN and a video transformer which are used in parallel to generate diverse positive pairs for contrastive learning. This allows the model to learn strong representations from such diverse yet meaningful pairs. Furthermore, we introduce a temporal self-supervised learning module able to predict an Edit distance explicitly between two video sequences in the temporal order. This enables the model to learn a rich temporal representation that compensates strongly to the video-level representation learned by the CACL. We evaluate our method on the tasks of video retrieval and action recognition on UCF101 and HMDB51 datasets, where our method achieves excellent performance, surpassing the state-of-the-art methods such as VideoMoCo and MoCo+BE by a large margin. The code is made available at https://github.com/guoshengcv/CACL.

Yi Wang, Yinan He, Yizhuo Li et al.

This paper introduces InternVid, a large-scale video-centric multimodal dataset that enables learning powerful and transferable video-text representations for multimodal understanding and generation. The InternVid dataset contains over 7 million videos lasting nearly 760K hours, yielding 234M video clips accompanied by detailed descriptions of total 4.1B words. Our core contribution is to develop a scalable approach to autonomously build a high-quality video-text dataset with large language models (LLM), thereby showcasing its efficacy in learning video-language representation at scale. Specifically, we utilize a multi-scale approach to generate video-related descriptions. Furthermore, we introduce ViCLIP, a video-text representation learning model based on ViT-L. Learned on InternVid via contrastive learning, this model demonstrates leading zero-shot action recognition and competitive video retrieval performance. Beyond basic video understanding tasks like recognition and retrieval, our dataset and model have broad applications. They are particularly beneficial for generating interleaved video-text data for learning a video-centric dialogue system, advancing video-to-text and text-to-video generation research. These proposed resources provide a tool for researchers and practitioners interested in multimodal video understanding and generation.

Lei Chen, Zhan Tong, Yibing Song et al.

The relation modeling between actors and scene context advances video action detection where the correlation of multiple actors makes their action recognition challenging. Existing studies model each actor and scene relation to improve action recognition. However, the scene variations and background interference limit the effectiveness of this relation modeling. In this paper, we propose to select actor-related scene context, rather than directly leverage raw video scenario, to improve relation modeling. We develop a Cycle Actor-Context Relation network (CycleACR) where there is a symmetric graph that models the actor and context relations in a bidirectional form. Our CycleACR consists of the Actor-to-Context Reorganization (A2C-R) that collects actor features for context feature reorganizations, and the Context-to-Actor Enhancement (C2A-E) that dynamically utilizes reorganized context features for actor feature enhancement. Compared to existing designs that focus on C2A-E, our CycleACR introduces A2C-R for a more effective relation modeling. This modeling advances our CycleACR to achieve state-of-the-art performance on two popular action detection datasets (i.e., AVA and UCF101-24). We also provide ablation studies and visualizations as well to show how our cycle actor-context relation modeling improves video action detection. Code is available at https://github.com/MCG-NJU/CycleACR.

Jintao Lin, Zhaoyang Liu, Wenhai Wang et al.

Video recognition in an open and dynamic world is quite challenging, as we need to handle different settings such as close-set, long-tail, few-shot and open-set. By leveraging semantic knowledge from noisy text descriptions crawled from the Internet, we focus on the general video recognition (GVR) problem of solving different recognition tasks within a unified framework. The core contribution of this paper is twofold. First, we build a comprehensive video recognition benchmark of Kinetics-GVR, including four sub-task datasets to cover the mentioned settings. To facilitate the research of GVR, we propose to utilize external textual knowledge from the Internet and provide multi-source text descriptions for all action classes. Second, inspired by the flexibility of language representation, we present a unified visual-linguistic framework (VLG) to solve the problem of GVR by an effective two-stage training paradigm. Our VLG is first pre-trained on video and language datasets to learn a shared feature space, and then devises a flexible bi-modal attention head to collaborate high-level semantic concepts under different settings. Extensive results show that our VLG obtains the state-of-the-art performance under four settings. The superior performance demonstrates the effectiveness and generalization ability of our proposed framework. We hope our work makes a step towards the general video recognition and could serve as a baseline for future research. The code and models will be available at https://github.com/MCG-NJU/VLG.

Jing Tan, Xiaotong Zhao, Xintian Shi et al.

Traditional temporal action detection (TAD) usually handles untrimmed videos with small number of action instances from a single label (e.g., ActivityNet, THUMOS). However, this setting might be unrealistic as different classes of actions often co-occur in practice. In this paper, we focus on the task of multi-label temporal action detection that aims to localize all action instances from a multi-label untrimmed video. Multi-label TAD is more challenging as it requires for fine-grained class discrimination within a single video and precise localization of the co-occurring instances. To mitigate this issue, we extend the sparse query-based detection paradigm from the traditional TAD and propose the multi-label TAD framework of PointTAD. Specifically, our PointTAD introduces a small set of learnable query points to represent the important frames of each action instance. This point-based representation provides a flexible mechanism to localize the discriminative frames at boundaries and as well the important frames inside the action. Moreover, we perform the action decoding process with the Multi-level Interactive Module to capture both point-level and instance-level action semantics. Finally, our PointTAD employs an end-to-end trainable framework simply based on RGB input for easy deployment. We evaluate our proposed method on two popular benchmarks and introduce the new metric of detection-mAP for multi-label TAD. Our model outperforms all previous methods by a large margin under the detection-mAP metric, and also achieves promising results under the segmentation-mAP metric. Code is available at https://github.com/MCG-NJU/PointTAD.

Jiahao Wang, Guo Chen, Yifei Huang et al.

Most existing forecasting systems are memory-based methods, which attempt to mimic human forecasting ability by employing various memory mechanisms and have progressed in temporal modeling for memory dependency. Nevertheless, an obvious weakness of this paradigm is that it can only model limited historical dependence and can not transcend the past. In this paper, we rethink the temporal dependence of event evolution and propose a novel memory-anticipation-based paradigm to model an entire temporal structure, including the past, present, and future. Based on this idea, we present Memory-and-Anticipation Transformer (MAT), a memory-anticipation-based approach, to address the online action detection and anticipation tasks. In addition, owing to the inherent superiority of MAT, it can process online action detection and anticipation tasks in a unified manner. The proposed MAT model is tested on four challenging benchmarks TVSeries, THUMOS'14, HDD, and EPIC-Kitchens-100, for online action detection and anticipation tasks, and it significantly outperforms all existing methods. Code is available at https://github.com/Echo0125/Memory-and-Anticipation-Transformer.

Tao Lu, Chunxu Liu, Youxin Chen et al.

Aggregating neighbor features is essential for point cloud classification. In the existing work, each point in the cloud may inevitably be selected as the neighbors of multiple aggregation centers, as all centers will gather neighbor features from the whole point cloud independently. Thus each point has to participate in the calculation repeatedly and generates redundant duplicates in the memory, leading to intensive computation costs and memory consumption. Meanwhile, to pursue higher accuracy, previous methods often rely on a complex local aggregator to extract fine geometric representation, which further slows down the classification pipeline. To address these issues, we propose a new local aggregator of linear complexity for point cloud classification, coined as APP. Specifically, we introduce an auxiliary container as an anchor to exchange features between the source point and the aggregating center. Each source point pushes its feature to only one auxiliary container, and each center point pulls features from only one auxiliary container. This avoids the re-computation issue of each source point. To facilitate the learning of the local structure of cloud point, we use an online normal estimation module to provide the explainable geometric information to enhance our APP modeling capability. Our built network is more efficient than all the previous baselines with a clear margin while still consuming a lower memory. Experiments on both synthetic and real datasets demonstrate that APP-Net reaches comparable accuracies to other networks. It can process more than 10,000 samples per second with less than 10GB of memory on a single GPU. We will release the code in https://github.com/MCG-NJU/APP-Net.

Yutao Cui, Chenkai Zeng, Xiaoyu Zhao et al.

Multi-object tracking in sports scenes plays a critical role in gathering players statistics, supporting further analysis, such as automatic tactical analysis. Yet existing MOT benchmarks cast little attention on the domain, limiting its development. In this work, we present a new large-scale multi-object tracking dataset in diverse sports scenes, coined as \emph{SportsMOT}, where all players on the court are supposed to be tracked. It consists of 240 video sequences, over 150K frames (almost 15\times MOT17) and over 1.6M bounding boxes (3\times MOT17) collected from 3 sports categories, including basketball, volleyball and football. Our dataset is characterized with two key properties: 1) fast and variable-speed motion and 2) similar yet distinguishable appearance. We expect SportsMOT to encourage the MOT trackers to promote in both motion-based association and appearance-based association. We benchmark several state-of-the-art trackers and reveal the key challenge of SportsMOT lies in object association. To alleviate the issue, we further propose a new multi-object tracking framework, termed as \emph{MixSort}, introducing a MixFormer-like structure as an auxiliary association model to prevailing tracking-by-detection trackers. By integrating the customized appearance-based association with the original motion-based association, MixSort achieves state-of-the-art performance on SportsMOT and MOT17. Based on MixSort, we give an in-depth analysis and provide some profound insights into SportsMOT. The dataset and code will be available at https://deeperaction.github.io/datasets/sportsmot.html.

Zikang Wang, Boyu Chen, Zhengrong Yue et al.

Recent advances in video understanding have been driven by MLLMs. But these MLLMs are good at analyzing short videos, while suffering from difficulties in understanding videos with a longer context. To address this difficulty, several agent methods have been proposed, using MLLMs as agents for retrieving extra contextual knowledge in a long video. However, most existing agents ignore the key fact that a long video is composed with multiple shots, i.e., to answer the user question from a long video, it is critical to deeply understand its relevant shots like human. Without such insight, these agents often mistakenly find redundant even noisy temporal context, restricting their capacity for long video understanding. To fill this gap, we propose VideoChat-A1, a novel long video agent paradigm. Different from the previous works, our VideoChat-A1 can deeply think with long videos, via a distinct chain-of-shot reasoning paradigm. More specifically, it can progressively select the relevant shots of user question, and look into these shots in a coarse-to-fine partition. By multi-modal reasoning along the shot chain, VideoChat-A1 can effectively mimic step-by-step human thinking process, allowing the interactive discovery of preferable temporal context for thoughtful understanding in long videos. Extensive experiments show that, VideoChat-A1 achieves the state-of-the-art performance on the mainstream long video QA benchmarks, e.g., it achieves 77.0 on VideoMME (w/ subs) and 70.1 on EgoSchema, outperforming its strong baselines (e.g., InternVL2.5-8B and InternVideo2.5-8B), by up to 10.1\% and 6.2\%. Compared to leading closed-source GPT-4o and Gemini 1.5 Pro, VideoChat-A1 offers competitive accuracy, but only with 7% input frames and 12% inference time on average. The code is available on https://github.com/SpXace/VideoChat-A1.

Yizhuo Li, Kunchang Li, Yinan He et al.

Building video-language foundation models is costly and difficult due to the redundant nature of video data and the lack of high-quality video-language datasets. In this paper, we propose an efficient framework to harvest video foundation models from image ones. Our method is intuitively simple by randomly dropping input video patches and masking out input text during the post-pretraining procedure. The patch dropping boosts the training efficiency significantly and text masking enforces the learning of cross-modal fusion. We conduct extensive experiments to validate the effectiveness of our method on a wide range of video-language downstream tasks including various zero-shot tasks, video question answering, and video-text retrieval. Despite its simplicity, our method achieves state-of-the-art performances, which are comparable to some heavily pretrained video foundation models. Our method is extremely efficient and can be trained in less than one day on 8 GPUs, requiring only WebVid-10M as pretraining data. We hope our method can serve as a simple yet strong counterpart for prevalent video foundation models, provide useful insights when building them, and make large pretrained models more accessible and sustainable. This is part of the InternVideo project \url{https://github.com/OpenGVLab/InternVideo}.

Haisong Liu, Yao Teng, Tao Lu et al.

Camera-based 3D object detection in BEV (Bird's Eye View) space has drawn great attention over the past few years. Dense detectors typically follow a two-stage pipeline by first constructing a dense BEV feature and then performing object detection in BEV space, which suffers from complex view transformations and high computation cost. On the other side, sparse detectors follow a query-based paradigm without explicit dense BEV feature construction, but achieve worse performance than the dense counterparts. In this paper, we find that the key to mitigate this performance gap is the adaptability of the detector in both BEV and image space. To achieve this goal, we propose SparseBEV, a fully sparse 3D object detector that outperforms the dense counterparts. SparseBEV contains three key designs, which are (1) scale-adaptive self attention to aggregate features with adaptive receptive field in BEV space, (2) adaptive spatio-temporal sampling to generate sampling locations under the guidance of queries, and (3) adaptive mixing to decode the sampled features with dynamic weights from the queries. On the test split of nuScenes, SparseBEV achieves the state-of-the-art performance of 67.5 NDS. On the val split, SparseBEV achieves 55.8 NDS while maintaining a real-time inference speed of 23.5 FPS. Code is available at https://github.com/MCG-NJU/SparseBEV.

Xinhao Li, Zhenpeng Huang, Jing Wang et al.

With the growth of high-quality data and advancement in visual pre-training paradigms, Video Foundation Models (VFMs) have made significant progress recently, demonstrating their remarkable performance on traditional video understanding benchmarks. However, the existing benchmarks (e.g. Kinetics) and their evaluation protocols are often limited by relatively poor diversity, high evaluation costs, and saturated performance metrics. In this paper, we build a comprehensive benchmark suite to address these issues, namely VideoEval. Specifically, we establish the Video Task Adaption Benchmark (VidTAB) and the Video Embedding Benchmark (VidEB) from two perspectives: evaluating the task adaptability of VFMs under few-shot conditions and assessing their representation power by directly applying to downstream tasks. With VideoEval, we conduct a large-scale study on 20 popular open-source vision foundation models. Our study reveals some insightful findings on VFMs: 1) overall, current VFMs exhibit weak generalization across diverse tasks, 2) increasing video data, whether labeled or weakly-labeled video-text pairs, does not necessarily improve task performance, 3) the effectiveness of some pre-training paradigms may not be fully validated in previous benchmarks, and 4) combining different pre-training paradigms can help improve the generalization capabilities. We believe this study serves as an important complement to the current evaluation for VFMs and offers valuable insights for the future research.

Jiaqi Tang, Zhaoyang Liu, Jing Tan et al.

Generic event boundary detection (GEBD) is an important yet challenging task in video understanding, which aims at detecting the moments where humans naturally perceive event boundaries. In this paper, we present a local context modeling and global boundary decoding approach for GEBD task. Local context modeling sub-network is proposed to perceive diverse patterns of generic event boundaries, and it generates powerful video representations and reliable boundary confidence. Based on them, global boundary decoding sub-network is exploited to decode event boundaries from a global view. Our proposed method achieves 85.13% F1-score on Kinetics-GEBD testing set, which achieves a more than 22% F1-score boost compared to the baseline method. The code is available at https://github.com/JackyTown/GEBD_Challenge_CVPR2022.

Zhiyu Zhao, Bingkun Huang, Sen Xing et al.

Self-supervised foundation models have shown great potential in computer vision thanks to the pre-training paradigm of masked autoencoding. Scale is a primary factor influencing the performance of these foundation models. However, these large foundation models often result in high computational cost. This paper focuses on pre-training relatively small vision transformer models that could be efficiently adapted to downstream tasks. Specifically, taking inspiration from knowledge distillation in model compression, we propose a new asymmetric masked distillation (AMD) framework for pre-training relatively small models with autoencoding. The core of AMD is to devise an asymmetric masking strategy, where the teacher model is enabled to see more context information with a lower masking ratio, while the student model is still equipped with a high masking ratio. We design customized multi-layer feature alignment between the teacher encoder and student encoder to regularize the pre-training of student MAE. To demonstrate the effectiveness and versatility of AMD, we apply it to both ImageMAE and VideoMAE for pre-training relatively small ViT models. AMD achieved 84.6% classification accuracy on IN1K using the ViT-B model. And AMD achieves 73.3% classification accuracy using the ViT-B model on the Something-in-Something V2 dataset, a 3.7% improvement over the original ViT-B model from VideoMAE. We also transfer AMD pre-trained models to downstream tasks and obtain consistent performance improvement over the original masked autoencoding. The code and models are available at https://github.com/MCG-NJU/AMD.

Chen Xu, Yuhan Zhu, Haocheng Shen et al.

Prompt learning has been designed as an alternative to fine-tuning for adapting Vision-language (V-L) models to the downstream tasks. Previous works mainly focus on text prompt while visual prompt works are limited for V-L models. The existing visual prompt methods endure either mediocre performance or unstable training process, indicating the difficulty of visual prompt learning. In this paper, we propose a new Progressive Visual Prompt (ProVP) structure to strengthen the interactions among prompts of different layers. More importantly, our ProVP could effectively propagate the image embeddings to deep layers and behave partially similar to an instance adaptive prompt method. To alleviate generalization deterioration, we further propose a new contrastive feature re-formation, which prevents the serious deviation of the prompted visual feature from the fixed CLIP visual feature distribution. Combining both, our method (ProVP-Ref) is evaluated on 11 image benchmark datasets and achieves 7/11 state-of-theart results on both few-shot and base-to-novel settings. To the best of our knowledge, we are the first to demonstrate the superior performance of visual prompts in V-L models to previous prompt-based methods in downstream tasks. Meanwhile, it implies that our ProVP-Ref shows the best capability to adapt and to generalize.

Fengyuan Shi, Ruopeng Gao, Weilin Huang et al.

Multimodal transformer exhibits high capacity and flexibility to align image and text for visual grounding. However, the existing encoder-only grounding framework (e.g., TransVG) suffers from heavy computation due to the self-attention operation with quadratic time complexity. To address this issue, we present a new multimodal transformer architecture, coined as Dynamic Mutilmodal DETR (Dynamic MDETR), by decoupling the whole grounding process into encoding and decoding phases. The key observation is that there exists high spatial redundancy in images. Thus, we devise a new dynamic multimodal transformer decoder by exploiting this sparsity prior to speed up the visual grounding process. Specifically, our dynamic decoder is composed of a 2D adaptive sampling module and a text guided decoding module. The sampling module aims to select these informative patches by predicting the offsets with respect to a reference point, while the decoding module works for extracting the grounded object information by performing cross attention between image features and text features. These two modules are stacked alternatively to gradually bridge the modality gap and iteratively refine the reference point of grounded object, eventually realizing the objective of visual grounding. Extensive experiments on five benchmarks demonstrate that our proposed Dynamic MDETR achieves competitive trade-offs between computation and accuracy. Notably, using only 9% feature points in the decoder, we can reduce ~44% GFLOPs of the multimodal transformer, but still get higher accuracy than the encoder-only counterpart. In addition, to verify its generalization ability and scale up our Dynamic MDETR, we build the first one-stage CLIP empowered visual grounding framework, and achieve the state-of-the-art performance on these benchmarks.

Bingkun Huang, Zhiyu Zhao, Guozhen Zhang et al.

Masked autoencoding has shown excellent performance on self-supervised video representation learning. Temporal redundancy has led to a high masking ratio and customized masking strategy in VideoMAE. In this paper, we aim to further improve the performance of video masked autoencoding by introducing a motion guided masking strategy. Our key insight is that motion is a general and unique prior in video, which should be taken into account during masked pre-training. Our motion guided masking explicitly incorporates motion information to build temporal consistent masking volume. Based on this masking volume, we can track the unmasked tokens in time and sample a set of temporal consistent cubes from videos. These temporal aligned unmasked tokens will further relieve the information leakage issue in time and encourage the MGMAE to learn more useful structure information. We implement our MGMAE with an online efficient optical flow estimator and backward masking map warping strategy. We perform experiments on the datasets of Something-Something V2 and Kinetics-400, demonstrating the superior performance of our MGMAE to the original VideoMAE. In addition, we provide the visualization analysis to illustrate that our MGMAE can sample temporal consistent cubes in a motion-adaptive manner for more effective video pre-training.

Tao Wu, Mengqi Cao, Ziteng Gao et al.

Traditional video action detectors typically adopt the two-stage pipeline, where a person detector is first employed to generate actor boxes and then 3D RoIAlign is used to extract actor-specific features for classification. This detection paradigm requires multi-stage training and inference, and cannot capture context information outside the bounding box. Recently, a few query-based action detectors are proposed to predict action instances in an end-to-end manner. However, they still lack adaptability in feature sampling and decoding, thus suffering from the issues of inferior performance or slower convergence. In this paper, we propose a new one-stage sparse action detector, termed STMixer. STMixer is based on two core designs. First, we present a query-based adaptive feature sampling module, which endows our STMixer with the flexibility of mining a set of discriminative features from the entire spatiotemporal domain. Second, we devise a dual-branch feature mixing module, which allows our STMixer to dynamically attend to and mix video features along the spatial and the temporal dimension respectively for better feature decoding. Coupling these two designs with a video backbone yields an efficient end-to-end action detector. Without bells and whistles, our STMixer obtains the state-of-the-art results on the datasets of AVA, UCF101-24, and JHMDB.

Tao Lu, Mulin Yu, Linning Xu et al.

Neural rendering methods have significantly advanced photo-realistic 3D scene rendering in various academic and industrial applications. The recent 3D Gaussian Splatting method has achieved the state-of-the-art rendering quality and speed combining the benefits of both primitive-based representations and volumetric representations. However, it often leads to heavily redundant Gaussians that try to fit every training view, neglecting the underlying scene geometry. Consequently, the resulting model becomes less robust to significant view changes, texture-less area and lighting effects. We introduce Scaffold-GS, which uses anchor points to distribute local 3D Gaussians, and predicts their attributes on-the-fly based on viewing direction and distance within the view frustum. Anchor growing and pruning strategies are developed based on the importance of neural Gaussians to reliably improve the scene coverage. We show that our method effectively reduces redundant Gaussians while delivering high-quality rendering. We also demonstrates an enhanced capability to accommodate scenes with varying levels-of-detail and view-dependent observations, without sacrificing the rendering speed.

Jing Tan, Yuhong Wang, Gangshan Wu et al.

Generic Boundary Detection (GBD) aims at locating the general boundaries that divide videos into semantically coherent and taxonomy-free units, and could serve as an important pre-processing step for long-form video understanding. Previous works often separately handle these different types of generic boundaries with specific designs of deep networks from simple CNN to LSTM. Instead, in this paper, we present Temporal Perceiver, a general architecture with Transformer, offering a unified solution to the detection of arbitrary generic boundaries, ranging from shot-level, event-level, to scene-level GBDs. The core design is to introduce a small set of latent feature queries as anchors to compress the redundant video input into a fixed dimension via cross-attention blocks. Thanks to this fixed number of latent units, it greatly reduces the quadratic complexity of attention operation to a linear form of input frames. Specifically, to explicitly leverage the temporal structure of videos, we construct two types of latent feature queries: boundary queries and context queries, which handle the semantic incoherence and coherence accordingly. Moreover, to guide the learning of latent feature queries, we propose an alignment loss on the cross-attention maps to explicitly encourage the boundary queries to attend on the top boundary candidates. Finally, we present a sparse detection head on the compressed representation, and directly output the final boundary detection results without any post-processing module. We test our Temporal Perceiver on a variety of GBD benchmarks. Our method obtains the state-of-the-art results on all benchmarks with RGB single-stream features: SoccerNet-v2 (81.9% avg-mAP), Kinetics-GEBD (86.0% avg-f1), TAPOS (73.2% avg-f1), MovieScenes (51.9% AP and 53.1% Miou) and MovieNet (53.3% AP and 53.2% Miou), demonstrating the generalization ability of our Temporal Perceiver.

Yue Feng, Jingjing Li, Qijia Lu et al.

The rapid advancement of AI-driven video generation has transformed content creation, while simultaneously increasing the risk of misinformation through localized manipulations in long-form videos. Existing video forensic methods predominantly operate on short, independent clips, and thus fail to capture realistic scenarios where AI-generated content is sparsely embedded within otherwise authentic footage. To bridge this gap, we formulate the task of Temporal AI-Generated Segment Localization and Explanation, which targets authenticity detection, temporal localization, and interpretable analysis of manipulated segments in untrimmed long videos. We further introduce TASLE, a large-scale benchmark comprising 12,472 untrimmed videos with diverse manipulation patterns and rich annotation signals, including temporal boundaries, authenticity labels, and segment-level rationales. In addition, we propose MSLoc, a coarse-to-fine forensic baseline that combines a boundary-sensitive proposal generation module for efficient long-video scanning with an MLLM-based refinement module for precise boundary localization and interpretable reasoning. Experiments validate the effectiveness of the proposed baseline, highlighting the importance of segment-level explainable forensics for long-form AI-generated video analysis. Our dataset and code are publicly available at https://debby-0527.github.io/TASLE.

Jiange Yang, Wenhui Tan, Chuhao Jin et al.

Improving the generalization capabilities of general-purpose robotic manipulation agents in the real world has long been a significant challenge. Existing approaches often rely on collecting large-scale robotic data which is costly and time-consuming, such as the RT-1 dataset. However, due to insufficient diversity of data, these approaches typically suffer from limiting their capability in open-domain scenarios with new objects and diverse environments. In this paper, we propose a novel paradigm that effectively leverages language-reasoning segmentation mask generated by internet-scale foundation models, to condition robot manipulation tasks. By integrating the mask modality, which incorporates semantic, geometric, and temporal correlation priors derived from vision foundation models, into the end-to-end policy model, our approach can effectively and robustly perceive object pose and enable sample-efficient generalization learning, including new object instances, semantic categories, and unseen backgrounds. We first introduce a series of foundation models to ground natural language demands across multiple tasks. Secondly, we develop a two-stream 2D policy model based on imitation learning, which processes raw images and object masks to predict robot actions with a local-global perception manner. Extensive realworld experiments conducted on a Franka Emika robot arm demonstrate the effectiveness of our proposed paradigm and policy architecture. Demos can be found in our submitted video, and more comprehensive ones can be found in link1 or link2.

Jun Zhang, Teng Wang, Yuying Ge et al.

This paper does not introduce a novel method but instead establishes a straightforward, incremental, yet essential baseline for video temporal grounding (VTG), a core capability in video understanding. While multimodal large language models (MLLMs) excel at various video understanding tasks, the recipes for optimizing them for VTG remain under-explored. In this paper, we present TimeLens, a systematic investigation into building MLLMs with strong VTG ability, along two primary dimensions: data quality and algorithmic design. We first expose critical quality issues in existing VTG benchmarks and introduce TimeLens-Bench, comprising meticulously re-annotated versions of three popular benchmarks with strict quality criteria. Our analysis reveals dramatic model re-rankings compared to legacy benchmarks, confirming the unreliability of prior evaluation standards. We also address noisy training data through an automated re-annotation pipeline, yielding TimeLens-100K, a large-scale, high-quality training dataset. Building on our data foundation, we conduct in-depth explorations of algorithmic design principles, yielding a series of meaningful insights and effective yet efficient practices. These include interleaved textual encoding for time representation, a thinking-free reinforcement learning with verifiable rewards (RLVR) approach as the training paradigm, and carefully designed recipes for RLVR training. These efforts culminate in TimeLens models, a family of MLLMs with state-of-the-art VTG performance among open-source models and even surpass proprietary models such as GPT-5 and Gemini-2.5-Flash. All codes, data, and models will be released to facilitate future research.

Yansong Shi, Qingsong Zhao, Tianxiang Jiang et al.

The rapid advancement of multimodal large language models has demonstrated impressive capabilities, yet nearly all operate in an offline paradigm, hindering real-time interactivity. Addressing this gap, we introduce the Real-tIme Video intERaction Bench (RIVER Bench), designed for evaluating online video comprehension. RIVER Bench introduces a novel framework comprising Retrospective Memory, Live-Perception, and Proactive Anticipation tasks, closely mimicking interactive dialogues rather than responding to entire videos at once. We conducted detailed annotations using videos from diverse sources and varying lengths, and precisely defined the real-time interactive format. Evaluations across various model categories reveal that while offline models perform well in single question-answering tasks, they struggle with real-time processing. Addressing the limitations of existing models in online video interaction, especially their deficiencies in long-term memory and future perception, we proposed a general improvement method that enables models to interact with users more flexibly in real time. We believe this work will significantly advance the development of real-time interactive video understanding models and inspire future research in this emerging field. Datasets and code are publicly available at https://github.com/OpenGVLab/RIVER.

Chen Xu, Yuhan Zhu, Guozhen Zhang et al.

Prompt learning has emerged as an efficient and effective approach for transferring foundational Vision-Language Models (e.g., CLIP) to downstream tasks. However, current methods tend to overfit to seen categories, thereby limiting their generalization ability for unseen classes. In this paper, we propose a new method, Decoupled Prompt Learning (DPL), which reformulates the attention in prompt learning to alleviate this problem. Specifically, we theoretically investigate the collaborative process between prompts and instances (i.e., image patches/text tokens) by reformulating the original self-attention into four separate sub-processes. Through detailed analysis, we observe that certain sub-processes can be strengthened to bolster robustness and generalizability by some approximation techniques. Furthermore, we introduce language-conditioned textual prompting based on decoupled attention to naturally preserve the generalization of text input. Our approach is flexible for both visual and textual modalities, making it easily extendable to multi-modal prompt learning. By combining the proposed techniques, our approach achieves state-of-the-art performance on three representative benchmarks encompassing 15 image recognition datasets, while maintaining parameter-efficient. Moreover, our DPL does not rely on any auxiliary regularization task or extra training data, further demonstrating its remarkable generalization ability.

Zhe Chen, Weiyun Wang, Yue Cao et al.

We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. In this work, we delve into the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations. Through extensive evaluations on a wide range of benchmarks, including multi-discipline reasoning, document understanding, multi-image / video understanding, real-world comprehension, multimodal hallucination detection, visual grounding, multilingual capabilities, and pure language processing, InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet. Notably, our model is the first open-source MLLMs to surpass 70% on the MMMU benchmark, achieving a 3.7-point improvement through Chain-of-Thought (CoT) reasoning and showcasing strong potential for test-time scaling. We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems. HuggingFace demo see https://huggingface.co/spaces/OpenGVLab/InternVL

Zhenpeng Huang, Jiaqi Li, Zihan Jia et al.

We present LongVPO, a novel two-stage Direct Preference Optimization framework that enables short-context vision-language models to robustly understand ultra-long videos without any long-video annotations. In Stage 1, we synthesize preference triples by anchoring questions to individual short clips, interleaving them with distractors, and applying visual-similarity and question-specificity filtering to mitigate positional bias and ensure unambiguous supervision. We also approximate the reference model's scoring over long contexts by evaluating only the anchor clip, reducing computational overhead. In Stage 2, we employ a recursive captioning pipeline on long videos to generate scene-level metadata, then use a large language model to craft multi-segment reasoning queries and dispreferred responses, aligning the model's preferences through multi-segment reasoning tasks. With only 16K synthetic examples and no costly human labels, LongVPO outperforms the state-of-the-art open-source models on multiple long-video benchmarks, while maintaining strong short-video performance (e.g., on MVBench), offering a scalable paradigm for efficient long-form video understanding.