Junwei Liang

Xiaoyu Zhu, Po-Yao Huang, Junwei Liang et al. · meta-ai

We study the problem of human action recognition using motion capture (MoCap) sequences. Unlike existing techniques that take multiple manual steps to derive standardized skeleton representations as model input, we propose a novel Spatial-Temporal Mesh Transformer (STMT) to directly model the mesh sequences. The model uses a hierarchical transformer with intra-frame off-set attention and inter-frame self-attention. The attention mechanism allows the model to freely attend between any two vertex patches to learn non-local relationships in the spatial-temporal domain. Masked vertex modeling and future frame prediction are used as two self-supervised tasks to fully activate the bi-directional and auto-regressive attention in our hierarchical transformer. The proposed method achieves state-of-the-art performance compared to skeleton-based and point-cloud-based models on common MoCap benchmarks. Code is available at https://github.com/zgzxy001/STMT.

Chengzhi Lin, Ancong Wu, Junwei Liang et al. · cmu, tencent-ai

Cross-modal retrieval between videos and texts has gained increasing research interest due to the rapid emergence of videos on the web. Generally, a video contains rich instance and event information and the query text only describes a part of the information. Thus, a video can correspond to multiple different text descriptions and queries. We call this phenomenon the ``Video-Text Correspondence Ambiguity'' problem. Current techniques mostly concentrate on mining local or multi-level alignment between contents of a video and text (\textit{e.g.}, object to entity and action to verb). It is difficult for these methods to alleviate the video-text correspondence ambiguity by describing a video using only one single feature, which is required to be matched with multiple different text features at the same time. To address this problem, we propose a Text-Adaptive Multiple Visual Prototype Matching model, which automatically captures multiple prototypes to describe a video by adaptive aggregation of video token features. Given a query text, the similarity is determined by the most similar prototype to find correspondence in the video, which is termed text-adaptive matching. To learn diverse prototypes for representing the rich information in videos, we propose a variance loss to encourage different prototypes to attend to different contents of the video. Our method outperforms state-of-the-art methods on four public video retrieval datasets.

Junwei Liang, Enwei Zhang, Jun Zhang et al. · cmu, tencent-ai

We study the task of robust feature representations, aiming to generalize well on multiple datasets for action recognition. We build our method on Transformers for its efficacy. Although we have witnessed great progress for video action recognition in the past decade, it remains challenging yet valuable how to train a single model that can perform well across multiple datasets. Here, we propose a novel multi-dataset training paradigm, MultiTrain, with the design of two new loss terms, namely informative loss and projection loss, aiming to learn robust representations for action recognition. In particular, the informative loss maximizes the expressiveness of the feature embedding while the projection loss for each dataset mines the intrinsic relations between classes across datasets. We verify the effectiveness of our method on five challenging datasets, Kinetics-400, Kinetics-700, Moments-in-Time, Activitynet and Something-something-v2 datasets. Extensive experimental results show that our method can consistently improve state-of-the-art performance. Code and models are released.

Xiaoyu Zhu, Hao Zhou, Pengfei Xing et al. · deepmind

In this paper, we investigate the use of diffusion models which are pre-trained on large-scale image-caption pairs for open-vocabulary 3D semantic understanding. We propose a novel method, namely Diff2Scene, which leverages frozen representations from text-image generative models, along with salient-aware and geometric-aware masks, for open-vocabulary 3D semantic segmentation and visual grounding tasks. Diff2Scene gets rid of any labeled 3D data and effectively identifies objects, appearances, materials, locations and their compositions in 3D scenes. We show that it outperforms competitive baselines and achieves significant improvements over state-of-the-art methods. In particular, Diff2Scene improves the state-of-the-art method on ScanNet200 by 12%.

Silvio Giancola, Anthony Cioppa, Adrien Deliège et al.

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet.

Teli Ma, Rong Li, Junwei Liang

With the success of Large Language Models (LLMs), many Generative Vision-Language Models (GVLMs) have been constructed via multimodal instruction tuning. However, the performance of GVLMs in multimodal compositional reasoning remains under-explored. In this paper, we examine both the evaluation metrics (VisualGPTScore, etc.) and current benchmarks for evaluating the compositionality of GVLMs. We identify the syntactical bias in current benchmarks, which is exploited by the linguistic capability of GVLMs. The bias renders VisualGPTScore an insufficient metric for assessing GVLMs. To combat this, we first introduce a SyntaxBias Score, leveraging LLMs to quantify such bias for mitigation. A challenging new task is subsequently added to evaluate the robustness of GVLMs against inherent inclination toward syntactical correctness. Using the bias-mitigated datasets and the new task, we propose a novel benchmark, namely SyntActically DE-biased benchmark (SADE). Our study provides an unbiased benchmark for the compositionality of GVLMs, facilitating future research in this direction (Code and dataset are available at https://github.com/TeleeMa/SADE).

Jinhui Ye, Junwei Liang · cmu, tencent-ai

This paper studies introducing viewpoint invariant feature representations in existing action recognition architecture. Despite significant progress in action recognition, efficiently handling geometric variations in large-scale datasets remains challenging. To tackle this problem, we propose a novel Spatial-Temporal Alignment Network (STAN), which explicitly learns geometric invariant representations for action recognition. Notably, the STAN model is light-weighted and generic, which could be plugged into existing action recognition models (e.g., MViTv2) with a low extra computational cost. We test our STAN model on widely-used datasets like UCF101 and HMDB51. The experimental results show that the STAN model can consistently improve the state-of-the-art models in action recognition tasks in trained-from-scratch settings.

Rong Li, ShiJie Li, Xieyuanli Chen et al.

LiDAR semantic segmentation plays a crucial role in enabling autonomous driving and robots to understand their surroundings accurately and robustly. A multitude of methods exist within this domain, including point-based, range-image-based, polar-coordinate-based, and hybrid strategies. Among these, range-image-based techniques have gained widespread adoption in practical applications due to their efficiency. However, they face a significant challenge known as the ``many-to-one'' problem caused by the range image's limited horizontal and vertical angular resolution. As a result, around 20% of the 3D points can be occluded. In this paper, we present TFNet, a range-image-based LiDAR semantic segmentation method that utilizes temporal information to address this issue. Specifically, we incorporate a temporal fusion layer to extract useful information from previous scans and integrate it with the current scan. We then design a max-voting-based post-processing technique to correct false predictions, particularly those caused by the ``many-to-one'' issue. We evaluated the approach on two benchmarks and demonstrated that the plug-in post-processing technique is generic and can be applied to various networks.

Yujin Tang, Jiaming Zhou, Xiang Pan et al.

Accurate precipitation forecasting is a vital challenge of societal importance. Though data-driven approaches have emerged as a widely used solution, solely relying on data-driven approaches has limitations in modeling the underlying physics, making accurate predictions difficult. We focus on the Numerical Weather Prediction (NWP) post-processing based precipitation forecasting task to couple Machine Learning techniques with traditional NWP. This task remains challenging due to the imbalanced precipitation data and complex relationships between multiple meteorological variables. To address these limitations, we introduce the \textbf{PostRainBench}, a comprehensive multi-variable NWP post-processing benchmark, and \textbf{CAMT}, a simple yet effective Channel Attention Enhanced Multi-task Learning framework with a specially designed weighted loss function. Extensive experimental results on the proposed benchmark show that our method outperforms state-of-the-art methods by 6.3\%, 4.7\%, and 26.8\% in rain CSI and improvements of 15.6\%, 17.4\%, and 31.8\% over NWP predictions in heavy rain CSI on respective datasets. Most notably, our model is the first deep learning-based method to outperform NWP approaches in heavy rain conditions. These results highlight the potential impact of our model in reducing the severe consequences of extreme rainfall events. Our datasets and code are available at https://github.com/yyyujintang/PostRainBench.

Changlin Li, Jiawei Zhang, Sihao Lin et al.

The rapid advancements in Large Vision Models (LVMs), such as Vision Transformers (ViTs) and diffusion models, have led to an increasing demand for computational resources, resulting in substantial financial and environmental costs. This growing challenge highlights the necessity of developing efficient training methods for LVMs. Progressive learning, a training strategy in which model capacity gradually increases during training, has shown potential in addressing these challenges. In this paper, we present an advanced automated progressive learning (AutoProg) framework for efficient training of LVMs. We begin by focusing on the pre-training of LVMs, using ViTs as a case study, and propose AutoProg-One, an AutoProg scheme featuring momentum growth (MoGrow) and a one-shot growth schedule search. Beyond pre-training, we extend our approach to tackle transfer learning and fine-tuning of LVMs. We expand the scope of AutoProg to cover a wider range of LVMs, including diffusion models. First, we introduce AutoProg-Zero, by enhancing the AutoProg framework with a novel zero-shot unfreezing schedule search, eliminating the need for one-shot supernet training. Second, we introduce a novel Unique Stage Identifier (SID) scheme to bridge the gap during network growth. These innovations, integrated with the core principles of AutoProg, offer a comprehensive solution for efficient training across various LVM scenarios. Extensive experiments show that AutoProg accelerates ViT pre-training by up to 1.85x on ImageNet and accelerates fine-tuning of diffusion models by up to 2.86x, with comparable or even higher performance. This work provides a robust and scalable approach to efficient training of LVMs, with potential applications in a wide range of vision tasks. Code: https://github.com/changlin31/AutoProg-Zero

Zeying Gong, Yujin Tang, Junwei Liang

Although the Transformer has been the dominant architecture for time series forecasting tasks in recent years, a fundamental challenge remains: the permutation-invariant self-attention mechanism within Transformers leads to a loss of temporal information. To tackle these challenges, we propose PatchMixer, a novel CNN-based model. It introduces a permutation-variant convolutional structure to preserve temporal information. Diverging from conventional CNNs in this field, which often employ multiple scales or numerous branches, our method relies exclusively on depthwise separable convolutions. This allows us to extract both local features and global correlations using a single-scale architecture. Furthermore, we employ dual forecasting heads encompassing linear and nonlinear components to better model future curve trends and details. Our experimental results on seven time-series forecasting benchmarks indicate that compared with the state-of-the-art method and the best-performing CNN, PatchMixer yields $3.9\%$ and $21.2\%$ relative improvements, respectively, while being 2-3x faster than the most advanced method.

Yujin Tang, Peijie Dong, Zhenheng Tang et al.

Combining CNNs or ViTs, with RNNs for spatiotemporal forecasting, has yielded unparalleled results in predicting temporal and spatial dynamics. However, modeling extensive global information remains a formidable challenge; CNNs are limited by their narrow receptive fields, and ViTs struggle with the intensive computational demands of their attention mechanisms. The emergence of recent Mamba-based architectures has been met with enthusiasm for their exceptional long-sequence modeling capabilities, surpassing established vision models in efficiency and accuracy, which motivates us to develop an innovative architecture tailored for spatiotemporal forecasting. In this paper, we propose the VMRNN cell, a new recurrent unit that integrates the strengths of Vision Mamba blocks with LSTM. We construct a network centered on VMRNN cells to tackle spatiotemporal prediction tasks effectively. Our extensive evaluations show that our proposed approach secures competitive results on a variety of tasks while maintaining a smaller model size. Our code is available at https://github.com/yyyujintang/VMRNN-PyTorch.

Jinhui Ye, Xing Wang, Wenxiang Jiao et al.

Gloss-free sign language translation (SLT) aims to develop well-performing SLT systems with no requirement for the costly gloss annotations, but currently still lags behind gloss-based approaches significantly. In this paper, we identify a representation density problem that could be a bottleneck in restricting the performance of gloss-free SLT. Specifically, the representation density problem describes that the visual representations of semantically distinct sign gestures tend to be closely packed together in feature space, which makes gloss-free methods struggle with distinguishing different sign gestures and suffer from a sharp performance drop. To address the representation density problem, we introduce a simple but effective contrastive learning strategy, namely SignCL, which encourages gloss-free models to learn more discriminative feature representation in a self-supervised manner. Our experiments demonstrate that the proposed SignCL can significantly reduce the representation density and improve performance across various translation frameworks. Specifically, SignCL achieves a significant improvement in BLEU score for the Sign Language Transformer and GFSLT-VLP on the CSL-Daily dataset by 39% and 46%, respectively, without any increase of model parameters. Compared to Sign2GPT, a state-of-the-art method based on large-scale pre-trained vision and language models, SignCL achieves better performance with only 35% of its parameters. Implementation and Checkpoints are available at https://github.com/JinhuiYE/SignCL.

Xinyu Sun, Lizhao Liu, Hongyan Zhi et al.

We study zero-shot instance navigation, in which the agent navigates to a specific object without using object annotations for training. Previous object navigation approaches apply the image-goal navigation (ImageNav) task (go to the location of an image) for pretraining, and transfer the agent to achieve object goals using a vision-language model. However, these approaches lead to issues of semantic neglect, where the model fails to learn meaningful semantic alignments. In this paper, we propose a Prioritized Semantic Learning (PSL) method to improve the semantic understanding ability of navigation agents. Specifically, a semantic-enhanced PSL agent is proposed and a prioritized semantic training strategy is introduced to select goal images that exhibit clear semantic supervision and relax the reward function from strict exact view matching. At inference time, a semantic expansion inference scheme is designed to preserve the same granularity level of the goal semantic as training. Furthermore, for the popular HM3D environment, we present an Instance Navigation (InstanceNav) task that requires going to a specific object instance with detailed descriptions, as opposed to the Object Navigation (ObjectNav) task where the goal is defined merely by the object category. Our PSL agent outperforms the previous state-of-the-art by 66% on zero-shot ObjectNav in terms of success rate and is also superior on the new InstanceNav task. Code will be released at https://github.com/XinyuSun/PSL-InstanceNav.

Jiaming Zhou, Junwei Liang, Kun-Yu Lin et al.

Zero-shot action recognition (ZSAR) aims to learn an alignment model between videos and class descriptions of seen actions that is transferable to unseen actions. The text queries (class descriptions) used in existing ZSAR works, however, are often short action names that fail to capture the rich semantics in the videos, leading to misalignment. With the intuition that video content descriptions (e.g., video captions) can provide rich contextual information of visual concepts in videos, we propose to utilize human annotated video descriptions to enrich the semantics of the class descriptions of each action. However, all existing action video description datasets are limited in terms of the number of actions, the semantics of video descriptions, etc. To this end, we collect a large-scale action video descriptions dataset named ActionHub, which covers a total of 1,211 common actions and provides 3.6 million action video descriptions. With the proposed ActionHub dataset, we further propose a novel Cross-modality and Cross-action Modeling (CoCo) framework for ZSAR, which consists of a Dual Cross-modality Alignment module and a Cross-action Invariance Mining module. Specifically, the Dual Cross-modality Alignment module utilizes both action labels and video descriptions from ActionHub to obtain rich class semantic features for feature alignment. The Cross-action Invariance Mining module exploits a cycle-reconstruction process between the class semantic feature spaces of seen actions and unseen actions, aiming to guide the model to learn cross-action invariant representations. Extensive experimental results demonstrate that our CoCo framework significantly outperforms the state-of-the-art on three popular ZSAR benchmarks (i.e., Kinetics-ZSAR, UCF101 and HMDB51) under two different learning protocols in ZSAR. We will release our code, models, and the proposed ActionHub dataset.

Jiaming Zhou, Hanjun Li, Kun-Yu Lin et al.

Developing end-to-end action recognition models on long videos is fundamental and crucial for long-video action understanding. Due to the unaffordable cost of end-to-end training on the whole long videos, existing works generally train models on short clips trimmed from long videos. However, this ``trimming-then-training'' practice requires action interval annotations for clip-level supervision, i.e., knowing which actions are trimmed into the clips. Unfortunately, collecting such annotations is very expensive and prevents model training at scale. To this end, this work aims to build a weakly supervised end-to-end framework for training recognition models on long videos, with only video-level action category labels. Without knowing the precise temporal locations of actions in long videos, our proposed weakly supervised framework, namely AdaptFocus, estimates where and how likely the actions will occur to adaptively focus on informative action clips for end-to-end training. The effectiveness of the proposed AdaptFocus framework is demonstrated on three long-video datasets. Furthermore, for downstream long-video tasks, our AdaptFocus framework provides a weakly supervised feature extraction pipeline for extracting more robust long-video features, such that the state-of-the-art methods on downstream tasks are significantly advanced. We will release the code and models.

Jinhui Ye, Jiaming Zhou, Hui Xiong et al.

Vision transformers have recently emerged as an effective alternative to convolutional networks for action recognition. However, vision transformers still struggle with geometric variations prevalent in video data. This paper proposes a novel approach, GeoDeformer, designed to capture the variations inherent in action video by integrating geometric comprehension directly into the ViT architecture. Specifically, at the core of GeoDeformer is the Geometric Deformation Predictor, a module designed to identify and quantify potential spatial and temporal geometric deformations within the given video. Spatial deformations adjust the geometry within individual frames, while temporal deformations capture the cross-frame geometric dynamics, reflecting motion and temporal progression. To demonstrate the effectiveness of our approach, we incorporate it into the established MViTv2 framework, replacing the standard self-attention blocks with GeoDeformer blocks. Our experiments at UCF101, HMDB51, and Mini-K200 achieve significant increases in both Top-1 and Top-5 accuracy, establishing new state-of-the-art results with only a marginal increase in computational cost. Additionally, visualizations affirm that GeoDeformer effectively manifests explicit geometric deformations and minimizes geometric variations. Codes and checkpoints will be released.

Jian Chen, Peilin Zhou, Yining Hua et al.

Accurate evaluation of financial question answering (QA) systems necessitates a comprehensive dataset encompassing diverse question types and contexts. However, current financial QA datasets lack scope diversity and question complexity. This work introduces FinTextQA, a novel dataset for long-form question answering (LFQA) in finance. FinTextQA comprises 1,262 high-quality, source-attributed QA pairs extracted and selected from finance textbooks and government agency websites.Moreover, we developed a Retrieval-Augmented Generation (RAG)-based LFQA system, comprising an embedder, retriever, reranker, and generator. A multi-faceted evaluation approach, including human ranking, automatic metrics, and GPT-4 scoring, was employed to benchmark the performance of different LFQA system configurations under heightened noisy conditions. The results indicate that: (1) Among all compared generators, Baichuan2-7B competes closely with GPT-3.5-turbo in accuracy score; (2) The most effective system configuration on our dataset involved setting the embedder, retriever, reranker, and generator as Ada2, Automated Merged Retrieval, Bge-Reranker-Base, and Baichuan2-7B, respectively; (3) models are less susceptible to noise after the length of contexts reaching a specific threshold.

Rong Li, Shijie Li, Lingdong Kong et al.

3D Visual Grounding (3DVG) aims to locate objects in 3D scenes based on textual descriptions, essential for applications like augmented reality and robotics. Traditional 3DVG approaches rely on annotated 3D datasets and predefined object categories, limiting scalability and adaptability. To overcome these limitations, we introduce SeeGround, a zero-shot 3DVG framework leveraging 2D Vision-Language Models (VLMs) trained on large-scale 2D data. SeeGround represents 3D scenes as a hybrid of query-aligned rendered images and spatially enriched text descriptions, bridging the gap between 3D data and 2D-VLMs input formats. We propose two modules: the Perspective Adaptation Module, which dynamically selects viewpoints for query-relevant image rendering, and the Fusion Alignment Module, which integrates 2D images with 3D spatial descriptions to enhance object localization. Extensive experiments on ScanRefer and Nr3D demonstrate that our approach outperforms existing zero-shot methods by large margins. Notably, we exceed weakly supervised methods and rival some fully supervised ones, outperforming previous SOTA by 7.7% on ScanRefer and 7.1% on Nr3D, showcasing its effectiveness in complex 3DVG tasks.

Teli Ma, Zifan Wang, Jiaming Zhou et al.

Inferring affordable (i.e., graspable) parts of arbitrary objects based on human specifications is essential for robots advancing toward open-vocabulary manipulation. Current grasp planners, however, are hindered by limited vision-language comprehension and time-consuming 3D radiance modeling, restricting real-time, open-vocabulary interactions with objects. To address these limitations, we propose GLOVER, a unified Generalizable Open-Vocabulary Affordance Reasoning framework, which fine-tunes the Large Language Models (LLMs) to predict the visual affordance of graspable object parts within RGB feature space. We compile a dataset of over 10,000 images from human-object interactions, annotated with unified visual and linguistic affordance labels, to enable multi-modal fine-tuning. GLOVER inherits world knowledge and common-sense reasoning from LLMs, facilitating more fine-grained object understanding and sophisticated tool-use reasoning. To enable effective real-world deployment, we present Affordance-Aware Grasping Estimation (AGE), a non-parametric grasp planner that aligns the gripper pose with a superquadric surface derived from affordance data. In evaluations across 30 table-top real-world scenes, GLOVER achieves success rates of 86.0% in part identification and 76.3% in grasping, with speeds approximately 29 times faster in affordance reasoning and 40 times faster in grasping pose estimation than the previous state-of-the-art. We also validate the generalization across embodiments, showing effectiveness in humanoid robots with dexterous hands.

Jiaming Zhou, Ke Ye, Jiayi Liu et al.

The generalization capabilities of vision-language-action (VLA) models to unseen tasks are crucial to achieving general-purpose robotic manipulation in open-world settings. However, the cross-task generalization capabilities of existing VLA models remain significantly underexplored. To address this gap, we introduce AGNOSTOS, a novel simulation benchmark designed to rigorously evaluate cross-task zero-shot generalization in manipulation. AGNOSTOS comprises 23 unseen manipulation tasks for testing, distinct from common training task distributions, and incorporates two levels of generalization difficulty to assess robustness. Our systematic evaluation reveals that current VLA models, despite being trained on diverse datasets, struggle to generalize effectively to these unseen tasks. To overcome this limitation, we propose Cross-Task In-Context Manipulation (X-ICM), a method that conditions large language models (LLMs) on in-context demonstrations from seen tasks to predict action sequences for unseen tasks. Additionally, we introduce a dynamics-guided sample selection strategy that identifies relevant demonstrations by capturing cross-task dynamics. On AGNOSTOS, X-ICM significantly improves cross-task zero-shot generalization performance over leading VLAs. We believe AGNOSTOS and X-ICM will serve as valuable tools for advancing general-purpose robotic manipulation.

Xiaoyu Zhu, Junwei Liang, Po-Yao Huang et al.

We study the problem of unsupervised domain adaptation for egocentric videos. We propose a transformer-based model to learn class-discriminative and domain-invariant feature representations. It consists of two novel designs. The first module is called Generative Adversarial Domain Alignment Network with the aim of learning domain-invariant representations. It simultaneously learns a mask generator and a domain-invariant encoder in an adversarial way. The domain-invariant encoder is trained to minimize the distance between the source and target domain. The masking generator, conversely, aims at producing challenging masks by maximizing the domain distance. The second is a Masked Consistency Learning module to learn class-discriminative representations. It enforces the prediction consistency between the masked target videos and their full forms. To better evaluate the effectiveness of domain adaptation methods, we construct a more challenging benchmark for egocentric videos, U-Ego4D. Our method achieves state-of-the-art performance on the Epic-Kitchen and the proposed U-Ego4D benchmark.

Dicong Qiu, Jiadi You, Zeying Gong et al.

We present the Semantics-aware Dataset and Benchmark Generation Pipeline for Open-vocabulary Object Navigation in Dynamic Scenes (SD-OVON). It utilizes pretraining multimodal foundation models to generate infinite unique photo-realistic scene variants that adhere to real-world semantics and daily commonsense for the training and the evaluation of navigation agents, accompanied with a plugin for generating object navigation task episodes compatible to the Habitat simulator. In addition, we offer two pre-generated object navigation task datasets, SD-OVON-3k and SD-OVON-10k, comprising respectively about 3k and 10k episodes of the open-vocabulary object navigation task, derived from the SD-OVON-Scenes dataset with 2.5k photo-realistic scans of real-world environments and the SD-OVON-Objects dataset with 0.9k manually inspected scanned and artist-created manipulatable object models. Unlike prior datasets limited to static environments, SD-OVON covers dynamic scenes and manipulatable objects, facilitating both real-to-sim and sim-to-real robotic applications. This approach enhances the realism of navigation tasks, the training and the evaluation of open-vocabulary object navigation agents in complex settings. To demonstrate the effectiveness of our pipeline and datasets, we propose two baselines and evaluate them along with state-of-the-art baselines on SD-OVON-3k. The datasets, benchmark and source code are publicly available.

Dicong Qiu, Wenzong Ma, Zhenfu Pan et al.

Open-Vocabulary Mobile Manipulation (OVMM) is a crucial capability for autonomous robots, especially when faced with the challenges posed by unknown and dynamic environments. This task requires robots to explore and build a semantic understanding of their surroundings, generate feasible plans to achieve manipulation goals, adapt to environmental changes, and comprehend natural language instructions from humans. To address these challenges, we propose a novel framework that leverages the zero-shot detection and grounded recognition capabilities of pretraining visual-language models (VLMs) combined with dense 3D entity reconstruction to build 3D semantic maps. Additionally, we utilize large language models (LLMs) for spatial region abstraction and online planning, incorporating human instructions and spatial semantic context. We have built a 10-DoF mobile manipulation robotic platform JSR-1 and demonstrated in real-world robot experiments that our proposed framework can effectively capture spatial semantics and process natural language user instructions for zero-shot OVMM tasks under dynamic environment settings, with an overall navigation and task success rate of 80.95% and 73.33% over 105 episodes, and better SFT and SPL by 157.18% and 19.53% respectively compared to the baseline. Furthermore, the framework is capable of replanning towards the next most probable candidate location based on the spatial semantic context derived from the 3D semantic map when initial plans fail, keeping an average success rate of 76.67%.

Jiaming Zhou, Teli Ma, Kun-Yu Lin et al.

Learning generalizable visual representations across different embodied environments is essential for effective robotic manipulation in real-world scenarios. However, the limited scale and diversity of robot demonstration data pose a significant challenge. Recent research has explored leveraging large-scale human activity data for pre-training, but the substantial morphological differences between humans and robots introduce a significant human-robot domain discrepancy, hindering the generalization of these models to downstream manipulation tasks. To overcome this, we propose a novel adaptation paradigm that leverages readily available paired human-robot video data to bridge the domain gap. Our method employs a human-robot contrastive alignment loss to align the semantics of human and robot videos, adapting pre-trained models to the robot domain in a parameter-efficient manner. Experiments on 20 simulated tasks across two different benchmarks and five real-world tasks demonstrate significant improvements. These results span both single-task and language-conditioned multi-task settings, evaluated using two different pre-trained models. Compared to existing pre-trained models, our adaptation method improves the average success rate by over 7% across multiple tasks on both simulated benchmarks and real-world evaluations.

Teli Ma, Jiaming Zhou, Zifan Wang et al.

Developing robots capable of executing various manipulation tasks, guided by natural language instructions and visual observations of intricate real-world environments, remains a significant challenge in robotics. Such robot agents need to understand linguistic commands and distinguish between the requirements of different tasks. In this work, we present Sigma-Agent, an end-to-end imitation learning agent for multi-task robotic manipulation. Sigma-Agent incorporates contrastive Imitation Learning (contrastive IL) modules to strengthen vision-language and current-future representations. An effective and efficient multi-view querying Transformer (MVQ-Former) for aggregating representative semantic information is introduced. Sigma-Agent shows substantial improvement over state-of-the-art methods under diverse settings in 18 RLBench tasks, surpassing RVT by an average of 5.2% and 5.9% in 10 and 100 demonstration training, respectively. Sigma-Agent also achieves 62% success rate with a single policy in 5 real-world manipulation tasks. The code will be released upon acceptance.

Junwei Liang, Liangliang Cao, Xuehan Xiong et al.

This paper studies how to introduce viewpoint-invariant feature representations that can help action recognition and detection. Although we have witnessed great progress of action recognition in the past decade, it remains challenging yet interesting how to efficiently model the geometric variations in large scale datasets. This paper proposes a novel Spatial-Temporal Alignment Network (STAN) that aims to learn geometric invariant representations for action recognition and action detection. The STAN model is very light-weighted and generic, which could be plugged into existing action recognition models like ResNet3D and the SlowFast with a very low extra computational cost. We test our STAN model extensively on AVA, Kinetics-400, AVA-Kinetics, Charades, and Charades-Ego datasets. The experimental results show that the STAN model can consistently improve the state of the arts in both action detection and action recognition tasks. We will release our data, models and code.

Junwei Liang

With the advancement in computer vision deep learning, systems now are able to analyze an unprecedented amount of rich visual information from videos to enable applications such as autonomous driving, socially-aware robot assistant and public safety monitoring. Deciphering human behaviors to predict their future paths/trajectories and what they would do from videos is important in these applications. However, human trajectory prediction still remains a challenging task, as scene semantics and human intent are difficult to model. Many systems do not provide high-level semantic attributes to reason about pedestrian future. This design hinders prediction performance in video data from diverse domains and unseen scenarios. To enable optimal future human behavioral forecasting, it is crucial for the system to be able to detect and analyze human activities as well as scene semantics, passing informative features to the subsequent prediction module for context understanding.

Xiaoyu Zhu, Junwei Liang, Alexander Hauptmann

In this paper, we study the problem of efficiently assessing building damage after natural disasters like hurricanes, floods or fires, through aerial video analysis. We make two main contributions. The first contribution is a new dataset, consisting of user-generated aerial videos from social media with annotations of instance-level building damage masks. This provides the first benchmark for quantitative evaluation of models to assess building damage using aerial videos. The second contribution is a new model, namely MSNet, which contains novel region proposal network designs and an unsupervised score refinement network for confidence score calibration in both bounding box and mask branches. We show that our model achieves state-of-the-art results compared to previous methods in our dataset. We will release our data, models and code.

Junwei Liang, Lu Jiang, Alexander Hauptmann

This paper studies the problem of predicting future trajectories of people in unseen cameras of novel scenarios and views. We approach this problem through the real-data-free setting in which the model is trained only on 3D simulation data and applied out-of-the-box to a wide variety of real cameras. We propose a novel approach to learn robust representation through augmenting the simulation training data such that the representation can better generalize to unseen real-world test data. The key idea is to mix the feature of the hardest camera view with the adversarial feature of the original view. We refer to our method as SimAug. We show that SimAug achieves promising results on three real-world benchmarks using zero real training data, and state-of-the-art performance in the Stanford Drone and the VIRAT/ActEV dataset when using in-domain training data.

Junwei Liang, Lu Jiang, Kevin Murphy et al.

This paper studies the problem of predicting the distribution over multiple possible future paths of people as they move through various visual scenes. We make two main contributions. The first contribution is a new dataset, created in a realistic 3D simulator, which is based on real world trajectory data, and then extrapolated by human annotators to achieve different latent goals. This provides the first benchmark for quantitative evaluation of the models to predict multi-future trajectories. The second contribution is a new model to generate multiple plausible future trajectories, which contains novel designs of using multi-scale location encodings and convolutional RNNs over graphs. We refer to our model as Multiverse. We show that our model achieves the best results on our dataset, as well as on the real-world VIRAT/ActEV dataset (which just contains one possible future).

Junwei Liang, Jay D. Aronson, Alexander Hauptmann

Every minute, hundreds of hours of video are uploaded to social media sites and the Internet from around the world. This material creates a visual record of the experiences of a significant percentage of humanity and can help illuminate how we live in the present moment. When properly analyzed, this video can also help analysts to reconstruct events of interest, including war crimes, human rights violations, and terrorist acts. Machine learning and computer vision can play a crucial role in this process. In this technical report, we describe the Video Event Reconstruction and Analysis (VERA) system. This new tool brings together a variety of capabilities we have developed over the past few years (including video synchronization and geolocation to order unstructured videos lacking metadata over time and space, and sound recognition algorithms) to enable the reconstruction and analysis of events captured on video. Among other uses, VERA enables the localization of a shooter from just a few videos that include the sound of gunshots. To demonstrate the efficacy of this suite of tools, we present the results of estimating the shooter's location of the Las Vegas Shooting in 2017 and show that VERA accurately predicts the shooter's location using only the first few gunshots. We then point out future directions that can help improve the system and further reduce unnecessary human labor in the process. All of the components of VERA run through a web interface that enables human-in-the-loop verification to ensure accurate estimations. All relevant source code, including the web interface and machine learning models, is freely available on Github. We hope that researchers and software developers will be inspired to improve and expand this system moving forward to better meet the needs of human rights and public safety.

Junwei Liang, Lu Jiang, Juan Carlos Niebles et al.

Deciphering human behaviors to predict their future paths/trajectories and what they would do from videos is important in many applications. Motivated by this idea, this paper studies predicting a pedestrian's future path jointly with future activities. We propose an end-to-end, multi-task learning system utilizing rich visual features about human behavioral information and interaction with their surroundings. To facilitate the training, the network is learned with an auxiliary task of predicting future location in which the activity will happen. Experimental results demonstrate our state-of-the-art performance over two public benchmarks on future trajectory prediction. Moreover, our method is able to produce meaningful future activity prediction in addition to the path. The result provides the first empirical evidence that joint modeling of paths and activities benefits future path prediction.

Junwei Liang, Lu Jiang, Liangliang Cao et al.

Recent insights on language and vision with neural networks have been successfully applied to simple single-image visual question answering. However, to tackle real-life question answering problems on multimedia collections such as personal photos, we have to look at whole collections with sequences of photos or videos. When answering questions from a large collection, a natural problem is to identify snippets to support the answer. In this paper, we describe a novel neural network called Focal Visual-Text Attention network (FVTA) for collective reasoning in visual question answering, where both visual and text sequence information such as images and text metadata are presented. FVTA introduces an end-to-end approach that makes use of a hierarchical process to dynamically determine what media and what time to focus on in the sequential data to answer the question. FVTA can not only answer the questions well but also provides the justifications which the system results are based upon to get the answers. FVTA achieves state-of-the-art performance on the MemexQA dataset and competitive results on the MovieQA dataset.

Ryota Hinami, Junwei Liang, Shin'ichi Satoh et al.

We tackle the problem of learning concept classifiers from videos on the web without using manually labeled data. Although metadata attached to videos (e.g., video titles, descriptions) can be of help collecting training data for the target concept, the collected data is often very noisy. The main challenge is therefore how to select good examples from noisy training data. Previous approaches firstly learn easy examples that are unlikely to be noise and then gradually learn more complex examples. However, hard examples that are much different from easy ones are never learned. In this paper, we propose an approach called multimodal co-training (MMCo) for selecting good examples from noisy training data. MMCo jointly learns classifiers for multiple modalities that complement each other to select good examples. Since MMCo selects examples by consensus of multimodal classifiers, a hard example for one modality can still be used as a training example by exploiting the power of the other modalities. The algorithm is very simple and easily implemented but yields consistent and significant boosts in example selection and classification performance on the FCVID and YouTube8M benchmarks.

Lu Jiang, Junwei Liang, Liangliang Cao et al.

This paper proposes a new task, MemexQA: given a collection of photos or videos from a user, the goal is to automatically answer questions that help users recover their memory about events captured in the collection. Towards solving the task, we 1) present the MemexQA dataset, a large, realistic multimodal dataset consisting of real personal photos and crowd-sourced questions/answers, 2) propose MemexNet, a unified, end-to-end trainable network architecture for image, text and video question answering. Experimental results on the MemexQA dataset demonstrate that MemexNet outperforms strong baselines and yields the state-of-the-art on this novel and challenging task. The promising results on TextQA and VideoQA suggest MemexNet's efficacy and scalability across various QA tasks.

Junwei Liang, Lu Jiang, Deyu Meng et al.

Learning video concept detectors automatically from the big but noisy web data with no additional manual annotations is a novel but challenging area in the multimedia and the machine learning community. A considerable amount of videos on the web are associated with rich but noisy contextual information, such as the title, which provides weak annotations or labels about the video content. To leverage the big noisy web labels, this paper proposes a novel method called WEbly-Labeled Learning (WELL), which is established on the state-of-the-art machine learning algorithm inspired by the learning process of human. WELL introduces a number of novel multi-modal approaches to incorporate meaningful prior knowledge called curriculum from the noisy web videos. To investigate this problem, we empirically study the curriculum constructed from the multi-modal features of the videos collected from YouTube and Flickr. The efficacy and the scalability of WELL have been extensively demonstrated on two public benchmarks, including the largest multimedia dataset and the largest manually-labeled video set. The comprehensive experimental results demonstrate that WELL outperforms state-of-the-art studies by a statically significant margin on learning concepts from noisy web video data. In addition, the results also verify that WELL is robust to the level of noisiness in the video data. Notably, WELL trained on sufficient noisy web labels is able to achieve a comparable accuracy to supervised learning methods trained on the clean manually-labeled data.