Lu Yuan

Haotian Zhang, Pengchuan Zhang, Xiaowei Hu et al. · microsoft-research

We present GLIPv2, a grounded VL understanding model, that serves both localization tasks (e.g., object detection, instance segmentation) and Vision-Language (VL) understanding tasks (e.g., VQA, image captioning). GLIPv2 elegantly unifies localization pre-training and Vision-Language Pre-training (VLP) with three pre-training tasks: phrase grounding as a VL reformulation of the detection task, region-word contrastive learning as a novel region-word level contrastive learning task, and the masked language modeling. This unification not only simplifies the previous multi-stage VLP procedure but also achieves mutual benefits between localization and understanding tasks. Experimental results show that a single GLIPv2 model (all model weights are shared) achieves near SoTA performance on various localization and understanding tasks. The model also shows (1) strong zero-shot and few-shot adaption performance on open-vocabulary object detection tasks and (2) superior grounding capability on VL understanding tasks. Code will be released at https://github.com/microsoft/GLIP.

Sheng Shen, Chunyuan Li, Xiaowei Hu et al. · berkeley, gatech

The new generation of state-of-the-art computer vision systems are trained from natural language supervision, ranging from simple object category names to descriptive captions. This form of supervision ensures high generality and usability of the learned visual models, due to the broad concept coverage achieved via large-scale data collection process. Alternatively, we argue that learning with external knowledge is a promising way which leverages a much more structured source of supervision and offers sample efficiency. We propose K-LITE, a simple strategy to leverage external knowledge for building transferable visual systems: In training, it enriches entities in text with WordNet and Wiktionary knowledge, leading to an efficient and scalable approach to learning image representations that uses knowledge about the visual concepts. In evaluation, the text is also augmented with external knowledge and then used to reference learned visual concepts (or describe new ones) to enable zero-shot and few-shot transfer of the pre-trained models. We study the performance of K-LITE on two important computer vision problems, image classification and object detection, benchmarking on 20 and 13 different existing datasets, respectively. The proposed knowledge-augmented models show significant improvement in transfer learning performance over existing methods. Our code is available at https://github.com/microsoft/klite.

Xueyan Zou, Zi-Yi Dou, Jianwei Yang et al. · microsoft-research

We present X-Decoder, a generalized decoding model that can predict pixel-level segmentation and language tokens seamlessly. X-Decodert takes as input two types of queries: (i) generic non-semantic queries and (ii) semantic queries induced from text inputs, to decode different pixel-level and token-level outputs in the same semantic space. With such a novel design, X-Decoder is the first work that provides a unified way to support all types of image segmentation and a variety of vision-language (VL) tasks. Further, our design enables seamless interactions across tasks at different granularities and brings mutual benefits by learning a common and rich pixel-level visual-semantic understanding space, without any pseudo-labeling. After pretraining on a mixed set of a limited amount of segmentation data and millions of image-text pairs, X-Decoder exhibits strong transferability to a wide range of downstream tasks in both zero-shot and finetuning settings. Notably, it achieves (1) state-of-the-art results on open-vocabulary segmentation and referring segmentation on eight datasets; (2) better or competitive finetuned performance to other generalist and specialist models on segmentation and VL tasks; and (3) flexibility for efficient finetuning and novel task composition (e.g., referring captioning and image editing). Code, demo, video, and visualization are available at https://x-decoder-vl.github.io.

Xiaoyi Dong, Jianmin Bao, Yinglin Zheng et al.

This paper presents a simple yet effective framework MaskCLIP, which incorporates a newly proposed masked self-distillation into contrastive language-image pretraining. The core idea of masked self-distillation is to distill representation from a full image to the representation predicted from a masked image. Such incorporation enjoys two vital benefits. First, masked self-distillation targets local patch representation learning, which is complementary to vision-language contrastive focusing on text-related representation. Second, masked self-distillation is also consistent with vision-language contrastive from the perspective of training objective as both utilize the visual encoder for feature aligning, and thus is able to learn local semantics getting indirect supervision from the language. We provide specially designed experiments with a comprehensive analysis to validate the two benefits. Symmetrically, we also introduce the local semantic supervision into the text branch, which further improves the pretraining performance. With extensive experiments, we show that MaskCLIP, when applied to various challenging downstream tasks, achieves superior results in linear probing, finetuning, and zero-shot performance with the guidance of the language encoder. Code will be release at \url{https://github.com/LightDXY/MaskCLIP}.

Jianwei Yang, Chunyuan Li, Xiyang Dai et al.

We propose focal modulation networks (FocalNets in short), where self-attention (SA) is completely replaced by a focal modulation mechanism for modeling token interactions in vision. Focal modulation comprises three components: (i) hierarchical contextualization, implemented using a stack of depth-wise convolutional layers, to encode visual contexts from short to long ranges, (ii) gated aggregation to selectively gather contexts for each query token based on its content, and (iii) element-wise modulation or affine transformation to inject the aggregated context into the query. Extensive experiments show FocalNets outperform the state-of-the-art SA counterparts (e.g., Swin and Focal Transformers) with similar computational costs on the tasks of image classification, object detection, and segmentation. Specifically, FocalNets with tiny and base size achieve 82.3% and 83.9% top-1 accuracy on ImageNet-1K. After pretrained on ImageNet-22K in 224 resolution, it attains 86.5% and 87.3% top-1 accuracy when finetuned with resolution 224 and 384, respectively. When transferred to downstream tasks, FocalNets exhibit clear superiority. For object detection with Mask R-CNN, FocalNet base trained with 1\times outperforms the Swin counterpart by 2.1 points and already surpasses Swin trained with 3\times schedule (49.0 v.s. 48.5). For semantic segmentation with UPerNet, FocalNet base at single-scale outperforms Swin by 2.4, and beats Swin at multi-scale (50.5 v.s. 49.7). Using large FocalNet and Mask2former, we achieve 58.5 mIoU for ADE20K semantic segmentation, and 57.9 PQ for COCO Panoptic Segmentation. Using huge FocalNet and DINO, we achieved 64.3 and 64.4 mAP on COCO minival and test-dev, respectively, establishing new SoTA on top of much larger attention-based models like Swinv2-G and BEIT-3. Code and checkpoints are available at https://github.com/microsoft/FocalNet.

Wengang Zhou, Weilun Wang, Jianmin Bao et al.

Denoising Diffusion Probabilistic Models (DDPMs) have achieved remarkable success in various image generation tasks compared with Generative Adversarial Nets (GANs). Recent work on semantic image synthesis mainly follows the de facto GAN-based approaches, which may lead to unsatisfactory quality or diversity of generated images. In this paper, we propose a novel framework based on DDPM for semantic image synthesis. Unlike previous conditional diffusion model directly feeds the semantic layout and noisy image as input to a U-Net structure, which may not fully leverage the information in the input semantic mask, our framework processes semantic layout and noisy image differently. It feeds noisy image to the encoder of the U-Net structure while the semantic layout to the decoder by multi-layer spatially-adaptive normalization operators. To further improve the generation quality and semantic interpretability in semantic image synthesis, we introduce the classifier-free guidance sampling strategy, which acknowledge the scores of an unconditional model for sampling process. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our proposed method, achieving state-of-the-art performance in terms of fidelity (FID) and diversity (LPIPS). Our code and pretrained models are available at https://github.com/WeilunWang/semantic-diffusion-model.

Kan Wu, Jinnian Zhang, Houwen Peng et al.

Vision transformer (ViT) recently has drawn great attention in computer vision due to its remarkable model capability. However, most prevailing ViT models suffer from huge number of parameters, restricting their applicability on devices with limited resources. To alleviate this issue, we propose TinyViT, a new family of tiny and efficient small vision transformers pretrained on large-scale datasets with our proposed fast distillation framework. The central idea is to transfer knowledge from large pretrained models to small ones, while enabling small models to get the dividends of massive pretraining data. More specifically, we apply distillation during pretraining for knowledge transfer. The logits of large teacher models are sparsified and stored in disk in advance to save the memory cost and computation overheads. The tiny student transformers are automatically scaled down from a large pretrained model with computation and parameter constraints. Comprehensive experiments demonstrate the efficacy of TinyViT. It achieves a top-1 accuracy of 84.8% on ImageNet-1k with only 21M parameters, being comparable to Swin-B pretrained on ImageNet-21k while using 4.2 times fewer parameters. Moreover, increasing image resolutions, TinyViT can reach 86.5% accuracy, being slightly better than Swin-L while using only 11% parameters. Last but not the least, we demonstrate a good transfer ability of TinyViT on various downstream tasks. Code and models are available at https://github.com/microsoft/Cream/tree/main/TinyViT.

Jianwei Yang, Chunyuan Li, Pengchuan Zhang et al.

Visual recognition is recently learned via either supervised learning on human-annotated image-label data or language-image contrastive learning with webly-crawled image-text pairs. While supervised learning may result in a more discriminative representation, language-image pretraining shows unprecedented zero-shot recognition capability, largely due to the different properties of data sources and learning objectives. In this work, we introduce a new formulation by combining the two data sources into a common image-text-label space. In this space, we propose a new learning paradigm, called Unified Contrastive Learning (UniCL) with a single learning objective to seamlessly prompt the synergy of two data types. Extensive experiments show that our UniCL is an effective way of learning semantically rich yet discriminative representations, universally for image recognition in zero-shot, linear-probe, fully finetuning and transfer learning scenarios. Particularly, it attains gains up to 9.2% and 14.5% in average on zero-shot recognition benchmarks over the language-image contrastive learning and supervised learning methods, respectively. In linear probe setting, it also boosts the performance over the two methods by 7.3% and 3.4%, respectively. Our study also indicates that UniCL stand-alone is a good learner on pure image-label data, rivaling the supervised learning methods across three image classification datasets and two types of vision backbones, ResNet and Swin Transformer. Code is available at https://github.com/microsoft/UniCL.

Qiankun Liu, Zhentao Tan, Dongdong Chen et al.

Transformers have achieved great success in pluralistic image inpainting recently. However, we find existing transformer based solutions regard each pixel as a token, thus suffer from information loss issue from two aspects: 1) They downsample the input image into much lower resolutions for efficiency consideration, incurring information loss and extra misalignment for the boundaries of masked regions. 2) They quantize $256^3$ RGB pixels to a small number (such as 512) of quantized pixels. The indices of quantized pixels are used as tokens for the inputs and prediction targets of transformer. Although an extra CNN network is used to upsample and refine the low-resolution results, it is difficult to retrieve the lost information back.To keep input information as much as possible, we propose a new transformer based framework "PUT". Specifically, to avoid input downsampling while maintaining the computation efficiency, we design a patch-based auto-encoder P-VQVAE, where the encoder converts the masked image into non-overlapped patch tokens and the decoder recovers the masked regions from inpainted tokens while keeping the unmasked regions unchanged. To eliminate the information loss caused by quantization, an Un-Quantized Transformer (UQ-Transformer) is applied, which directly takes the features from P-VQVAE encoder as input without quantization and regards the quantized tokens only as prediction targets. Extensive experiments show that PUT greatly outperforms state-of-the-art methods on image fidelity, especially for large masked regions and complex large-scale datasets. Code is available at https://github.com/liuqk3/PUT

Xiaoyi Dong, Jianmin Bao, Ting Zhang et al.

We propose bootstrapped masked autoencoders (BootMAE), a new approach for vision BERT pretraining. BootMAE improves the original masked autoencoders (MAE) with two core designs: 1) momentum encoder that provides online feature as extra BERT prediction targets; 2) target-aware decoder that tries to reduce the pressure on the encoder to memorize target-specific information in BERT pretraining. The first design is motivated by the observation that using a pretrained MAE to extract the features as the BERT prediction target for masked tokens can achieve better pretraining performance. Therefore, we add a momentum encoder in parallel with the original MAE encoder, which bootstraps the pretraining performance by using its own representation as the BERT prediction target. In the second design, we introduce target-specific information (e.g., pixel values of unmasked patches) from the encoder directly to the decoder to reduce the pressure on the encoder of memorizing the target-specific information. Thus, the encoder focuses on semantic modeling, which is the goal of BERT pretraining, and does not need to waste its capacity in memorizing the information of unmasked tokens related to the prediction target. Through extensive experiments, our BootMAE achieves $84.2\%$ Top-1 accuracy on ImageNet-1K with ViT-B backbone, outperforming MAE by $+0.8\%$ under the same pre-training epochs. BootMAE also gets $+1.0$ mIoU improvements on semantic segmentation on ADE20K and $+1.3$ box AP, $+1.4$ mask AP improvement on object detection and segmentation on COCO dataset. Code is released at https://github.com/LightDXY/BootMAE.

Mingyu Ding, Bin Xiao, Noel Codella et al.

In this work, we introduce Dual Attention Vision Transformers (DaViT), a simple yet effective vision transformer architecture that is able to capture global context while maintaining computational efficiency. We propose approaching the problem from an orthogonal angle: exploiting self-attention mechanisms with both "spatial tokens" and "channel tokens". With spatial tokens, the spatial dimension defines the token scope, and the channel dimension defines the token feature dimension. With channel tokens, we have the inverse: the channel dimension defines the token scope, and the spatial dimension defines the token feature dimension. We further group tokens along the sequence direction for both spatial and channel tokens to maintain the linear complexity of the entire model. We show that these two self-attentions complement each other: (i) since each channel token contains an abstract representation of the entire image, the channel attention naturally captures global interactions and representations by taking all spatial positions into account when computing attention scores between channels; (ii) the spatial attention refines the local representations by performing fine-grained interactions across spatial locations, which in turn helps the global information modeling in channel attention. Extensive experiments show our DaViT achieves state-of-the-art performance on four different tasks with efficient computations. Without extra data, DaViT-Tiny, DaViT-Small, and DaViT-Base achieve 82.8%, 84.2%, and 84.6% top-1 accuracy on ImageNet-1K with 28.3M, 49.7M, and 87.9M parameters, respectively. When we further scale up DaViT with 1.5B weakly supervised image and text pairs, DaViT-Gaint reaches 90.4% top-1 accuracy on ImageNet-1K. Code is available at https://github.com/dingmyu/davit.

Rui Wang, Dongdong Chen, Zuxuan Wu et al.

Benefiting from masked visual modeling, self-supervised video representation learning has achieved remarkable progress. However, existing methods focus on learning representations from scratch through reconstructing low-level features like raw pixel RGB values. In this paper, we propose masked video distillation (MVD), a simple yet effective two-stage masked feature modeling framework for video representation learning: firstly we pretrain an image (or video) model by recovering low-level features of masked patches, then we use the resulting features as targets for masked feature modeling. For the choice of teacher models, we observe that students taught by video teachers perform better on temporally-heavy video tasks, while image teachers transfer stronger spatial representations for spatially-heavy video tasks. Visualization analysis also indicates different teachers produce different learned patterns for students. Motivated by this observation, we design a spatial-temporal co-teaching method for MVD. Specifically, we distill student models from both video teachers and image teachers by masked feature modeling. Extensive experimental results demonstrate that video transformers pretrained with spatial-temporal co-teaching outperform models distilled with a single teacher on a multitude of video datasets. Our MVD with vanilla ViT achieves state-of-the-art performance compared with previous supervised or self-supervised methods on several challenging video downstream tasks. For example, with the ViT-Large model, our MVD achieves 86.4% and 76.7% Top-1 accuracy on Kinetics-400 and Something-Something-v2, outperforming VideoMAE by 1.2% and 2.4% respectively. When a larger ViT-Huge model is adopted, MVD achieves the state-of-the-art performance with 77.3% Top-1 accuracy on Something-Something-v2 and 41.1 mAP on AVA v2.2. Code will be available at \url{https://github.com/ruiwang2021/mvd}.

Wan-Cyuan Fan, Yen-Chun Chen, Dongdong Chen et al.

Diffusion models (DMs) have shown great potential for high-quality image synthesis. However, when it comes to producing images with complex scenes, how to properly describe both image global structures and object details remains a challenging task. In this paper, we present Frido, a Feature Pyramid Diffusion model performing a multi-scale coarse-to-fine denoising process for image synthesis. Our model decomposes an input image into scale-dependent vector quantized features, followed by a coarse-to-fine gating for producing image output. During the above multi-scale representation learning stage, additional input conditions like text, scene graph, or image layout can be further exploited. Thus, Frido can be also applied for conditional or cross-modality image synthesis. We conduct extensive experiments over various unconditioned and conditional image generation tasks, ranging from text-to-image synthesis, layout-to-image, scene-graph-to-image, to label-to-image. More specifically, we achieved state-of-the-art FID scores on five benchmarks, namely layout-to-image on COCO and OpenImages, scene-graph-to-image on COCO and Visual Genome, and label-to-image on COCO. Code is available at https://github.com/davidhalladay/Frido.

Yuanze Lin, Yujia Xie, Dongdong Chen et al.

This paper revisits visual representation in knowledge-based visual question answering (VQA) and demonstrates that using regional information in a better way can significantly improve the performance. While visual representation is extensively studied in traditional VQA, it is under-explored in knowledge-based VQA even though these two tasks share the common spirit, i.e., rely on visual input to answer the question. Specifically, we observe that in most state-of-the-art knowledge-based VQA methods: 1) visual features are extracted either from the whole image or in a sliding window manner for retrieving knowledge, and the important relationship within/among object regions is neglected; 2) visual features are not well utilized in the final answering model, which is counter-intuitive to some extent. Based on these observations, we propose a new knowledge-based VQA method REVIVE, which tries to utilize the explicit information of object regions not only in the knowledge retrieval stage but also in the answering model. The key motivation is that object regions and inherent relationship are important for knowledge-based VQA. We perform extensive experiments on the standard OK-VQA dataset and achieve new state-of-the-art performance, i.e., 58.0% accuracy, surpassing previous state-of-the-art method by a large margin (+3.6%). We also conduct detailed analysis and show the necessity of regional information in different framework components for knowledge-based VQA. Code is publicly available at https://github.com/yzleroy/REVIVE.

Xiaoyi Dong, Jianmin Bao, Ting Zhang et al.

Recent studies have shown that CLIP has achieved remarkable success in performing zero-shot inference while its fine-tuning performance is not satisfactory. In this paper, we identify that fine-tuning performance is significantly impacted by hyper-parameter choices. We examine various key hyper-parameters and empirically evaluate their impact in fine-tuning CLIP for classification tasks through a comprehensive study. We find that the fine-tuning performance of CLIP is substantially underestimated. Equipped with hyper-parameter refinement, we demonstrate CLIP itself is better or at least competitive in fine-tuning compared with large-scale supervised pre-training approaches or latest works that use CLIP as prediction targets in Masked Image Modeling. Specifically, CLIP ViT-Base/16 and CLIP ViT-Large/14 can achieve 85.7%,88.0% finetuning Top-1 accuracy on the ImageNet-1K dataset . These observations challenge the conventional conclusion that CLIP is not suitable for fine-tuning, and motivate us to rethink recently proposed improvements based on CLIP. We will release our code publicly at \url{https://github.com/LightDXY/FT-CLIP}.

Hanqing Zhao, Dianmo Sheng, Jianmin Bao et al.

Copy-Paste is a simple and effective data augmentation strategy for instance segmentation. By randomly pasting object instances onto new background images, it creates new training data for free and significantly boosts the segmentation performance, especially for rare object categories. Although diverse, high-quality object instances used in Copy-Paste result in more performance gain, previous works utilize object instances either from human-annotated instance segmentation datasets or rendered from 3D object models, and both approaches are too expensive to scale up to obtain good diversity. In this paper, we revisit Copy-Paste at scale with the power of newly emerged zero-shot recognition models (e.g., CLIP) and text2image models (e.g., StableDiffusion). We demonstrate for the first time that using a text2image model to generate images or zero-shot recognition model to filter noisily crawled images for different object categories is a feasible way to make Copy-Paste truly scalable. To make such success happen, we design a data acquisition and processing framework, dubbed ``X-Paste", upon which a systematic study is conducted. On the LVIS dataset, X-Paste provides impressive improvements over the strong baseline CenterNet2 with Swin-L as the backbone. Specifically, it archives +2.6 box AP and +2.1 mask AP gains on all classes and even more significant gains with +6.8 box AP, +6.5 mask AP on long-tail classes. Our code and models are available at https://github.com/yoctta/XPaste.

Zixin Zhu, Xuelu Feng, Dongdong Chen et al.

StableDiffusion is a revolutionary text-to-image generator that is causing a stir in the world of image generation and editing. Unlike traditional methods that learn a diffusion model in pixel space, StableDiffusion learns a diffusion model in the latent space via a VQGAN, ensuring both efficiency and quality. It not only supports image generation tasks, but also enables image editing for real images, such as image inpainting and local editing. However, we have observed that the vanilla VQGAN used in StableDiffusion leads to significant information loss, causing distortion artifacts even in non-edited image regions. To this end, we propose a new asymmetric VQGAN with two simple designs. Firstly, in addition to the input from the encoder, the decoder contains a conditional branch that incorporates information from task-specific priors, such as the unmasked image region in inpainting. Secondly, the decoder is much heavier than the encoder, allowing for more detailed recovery while only slightly increasing the total inference cost. The training cost of our asymmetric VQGAN is cheap, and we only need to retrain a new asymmetric decoder while keeping the vanilla VQGAN encoder and StableDiffusion unchanged. Our asymmetric VQGAN can be widely used in StableDiffusion-based inpainting and local editing methods. Extensive experiments demonstrate that it can significantly improve the inpainting and editing performance, while maintaining the original text-to-image capability. The code is available at \url{https://github.com/buxiangzhiren/Asymmetric_VQGAN}.

Qinhong Yang, Dongdong Chen, Zhentao Tan et al.

This paper introduces a new large-scale image restoration dataset, called HQ-50K, which contains 50,000 high-quality images with rich texture details and semantic diversity. We analyze existing image restoration datasets from five different perspectives, including data scale, resolution, compression rates, texture details, and semantic coverage. However, we find that all of these datasets are deficient in some aspects. In contrast, HQ-50K considers all of these five aspects during the data curation process and meets all requirements. We also present a new Degradation-Aware Mixture of Expert (DAMoE) model, which enables a single model to handle multiple corruption types and unknown levels. Our extensive experiments demonstrate that HQ-50K consistently improves the performance on various image restoration tasks, such as super-resolution, denoising, dejpeg, and deraining. Furthermore, our proposed DAMoE, trained on our \dataset, outperforms existing state-of-the-art unified models designed for multiple restoration tasks and levels. The dataset and code are available at \url{https://github.com/littleYaang/HQ-50K}.

Zhecan Wang, Noel Codella, Yen-Chun Chen et al. · microsoft-research

Cross-modal encoders for vision-language (VL) tasks are often pretrained with carefully curated vision-language datasets. While these datasets reach an order of 10 million samples, the labor cost is prohibitive to scale further. Conversely, unimodal encoders are pretrained with simpler annotations that are less cost-prohibitive, achieving scales of hundreds of millions to billions. As a result, unimodal encoders have achieved state-of-art (SOTA) on many downstream tasks. However, challenges remain when applying to VL tasks. The pretraining data is not optimal for cross-modal architectures and requires heavy computational resources. In addition, unimodal architectures lack cross-modal interactions that have demonstrated significant benefits for VL tasks. Therefore, how to best leverage pretrained unimodal encoders for VL tasks is still an area of active research. In this work, we propose a method to leverage unimodal vision and text encoders for VL tasks that augment existing VL approaches while conserving computational complexity. Specifically, we propose Multimodal Adaptive Distillation (MAD), which adaptively distills useful knowledge from pretrained encoders to cross-modal VL encoders. Second, to better capture nuanced impacts on VL task performance, we introduce an evaluation protocol that includes Visual Commonsense Reasoning (VCR), Visual Entailment (SNLI-VE), and Visual Question Answering (VQA), across a variety of data constraints and conditions of domain shift. Experiments demonstrate that MAD leads to consistent gains in the low-shot, domain-shifted, and fully-supervised conditions on VCR, SNLI-VE, and VQA, achieving SOTA performance on VCR compared to other single models pretrained with image-text data. Finally, MAD outperforms concurrent works utilizing pretrained vision encoder from CLIP. Code will be made available.

Ziyi Yang, Yuwei Fang, Chenguang Zhu et al. · gatech, stanford

Human intelligence is multimodal; we integrate visual, linguistic, and acoustic signals to maintain a holistic worldview. Most current pretraining methods, however, are limited to one or two modalities. We present i-Code, a self-supervised pretraining framework where users may flexibly combine the modalities of vision, speech, and language into unified and general-purpose vector representations. In this framework, data from each modality are first given to pretrained single-modality encoders. The encoder outputs are then integrated with a multimodal fusion network, which uses novel attention mechanisms and other architectural innovations to effectively combine information from the different modalities. The entire system is pretrained end-to-end with new objectives including masked modality unit modeling and cross-modality contrastive learning. Unlike previous research using only video for pretraining, the i-Code framework can dynamically process single, dual, and triple-modality data during training and inference, flexibly projecting different combinations of modalities into a single representation space. Experimental results demonstrate how i-Code can outperform state-of-the-art techniques on five video understanding tasks and the GLUE NLP benchmark, improving by as much as 11% and demonstrating the power of integrative multimodal pretraining.

Kan Wu, Houwen Peng, Zhenghong Zhou et al.

In this paper, we propose a novel cross-modal distillation method, called TinyCLIP, for large-scale language-image pre-trained models. The method introduces two core techniques: affinity mimicking and weight inheritance. Affinity mimicking explores the interaction between modalities during distillation, enabling student models to mimic teachers' behavior of learning cross-modal feature alignment in a visual-linguistic affinity space. Weight inheritance transmits the pre-trained weights from the teacher models to their student counterparts to improve distillation efficiency. Moreover, we extend the method into a multi-stage progressive distillation to mitigate the loss of informative weights during extreme compression. Comprehensive experiments demonstrate the efficacy of TinyCLIP, showing that it can reduce the size of the pre-trained CLIP ViT-B/32 by 50%, while maintaining comparable zero-shot performance. While aiming for comparable performance, distillation with weight inheritance can speed up the training by 1.4 - 7.8 $\times$ compared to training from scratch. Moreover, our TinyCLIP ViT-8M/16, trained on YFCC-15M, achieves an impressive zero-shot top-1 accuracy of 41.1% on ImageNet, surpassing the original CLIP ViT-B/16 by 3.5% while utilizing only 8.9% parameters. Finally, we demonstrate the good transferability of TinyCLIP in various downstream tasks. Code and models will be open-sourced at https://aka.ms/tinyclip.

Haoxuan You, Luowei Zhou, Bin Xiao et al.

Large-scale multi-modal contrastive pre-training has demonstrated great utility to learn transferable features for a range of downstream tasks by mapping multiple modalities into a shared embedding space. Typically, this has employed separate encoders for each modality. However, recent work suggests that transformers can support learning across multiple modalities and allow knowledge sharing. Inspired by this, we investigate a variety of Modality-Shared Contrastive Language-Image Pre-training (MS-CLIP) frameworks. More specifically, we question how many parameters of a transformer model can be shared across modalities during contrastive pre-training, and rigorously examine architectural design choices that position the proportion of parameters shared along a spectrum. In studied conditions, we observe that a mostly unified encoder for vision and language signals outperforms all other variations that separate more parameters. Additionally, we find that light-weight modality-specific parallel modules further improve performance. Experimental results show that the proposed MS-CLIP approach outperforms vanilla CLIP by up to 13\% relative in zero-shot ImageNet classification (pre-trained on YFCC-100M), while simultaneously supporting a reduction of parameters. In addition, our approach outperforms vanilla CLIP by 1.6 points in linear probing on a collection of 24 downstream vision tasks. Furthermore, we discover that sharing parameters leads to semantic concepts from different modalities being encoded more closely in the embedding space, facilitating the transferring of common semantic structure (e.g., attention patterns) from language to vision. Code is available at \href{https://github.com/Hxyou/MSCLIP}{URL}.

Shuquan Ye, Yujia Xie, Dongdong Chen et al.

This paper focuses on analyzing and improving the commonsense ability of recent popular vision-language (VL) models. Despite the great success, we observe that existing VL-models still lack commonsense knowledge/reasoning ability (e.g., "Lemons are sour"), which is a vital component towards artificial general intelligence. Through our analysis, we find one important reason is that existing large-scale VL datasets do not contain much commonsense knowledge, which motivates us to improve the commonsense of VL-models from the data perspective. Rather than collecting a new VL training dataset, we propose a more scalable strategy, i.e., "Data Augmentation with kNowledge graph linearization for CommonsensE capability" (DANCE). It can be viewed as one type of data augmentation technique, which can inject commonsense knowledge into existing VL datasets on the fly during training. More specifically, we leverage the commonsense knowledge graph (e.g., ConceptNet) and create variants of text description in VL datasets via bidirectional sub-graph sequentialization. For better commonsense evaluation, we further propose the first retrieval-based commonsense diagnostic benchmark. By conducting extensive experiments on some representative VL-models, we demonstrate that our DANCE technique is able to significantly improve the commonsense ability while maintaining the performance on vanilla retrieval tasks. The code and data are available at https://github.com/pleaseconnectwifi/DANCE

Qidong Huang, Xiaoyi Dong, Dongdong Chen et al.

In this paper, we investigate the adversarial robustness of vision transformers that are equipped with BERT pretraining (e.g., BEiT, MAE). A surprising observation is that MAE has significantly worse adversarial robustness than other BERT pretraining methods. This observation drives us to rethink the basic differences between these BERT pretraining methods and how these differences affect the robustness against adversarial perturbations. Our empirical analysis reveals that the adversarial robustness of BERT pretraining is highly related to the reconstruction target, i.e., predicting the raw pixels of masked image patches will degrade more adversarial robustness of the model than predicting the semantic context, since it guides the model to concentrate more on medium-/high-frequency components of images. Based on our analysis, we provide a simple yet effective way to boost the adversarial robustness of MAE. The basic idea is using the dataset-extracted domain knowledge to occupy the medium-/high-frequency of images, thus narrowing the optimization space of adversarial perturbations. Specifically, we group the distribution of pretraining data and optimize a set of cluster-specific visual prompts on frequency domain. These prompts are incorporated with input images through prototype-based prompt selection during test period. Extensive evaluation shows that our method clearly boost MAE's adversarial robustness while maintaining its clean performance on ImageNet-1k classification. Our code is available at: https://github.com/shikiw/RobustMAE.

Lingchen Meng, Xiyang Dai, Jianwei Yang et al.

Long-tailed object detection (LTOD) aims to handle the extreme data imbalance in real-world datasets, where many tail classes have scarce instances. One popular strategy is to explore extra data with image-level labels, yet it produces limited results due to (1) semantic ambiguity -- an image-level label only captures a salient part of the image, ignoring the remaining rich semantics within the image; and (2) location sensitivity -- the label highly depends on the locations and crops of the original image, which may change after data transformations like random cropping. To remedy this, we propose RichSem, a simple but effective method, which is robust to learn rich semantics from coarse locations without the need of accurate bounding boxes. RichSem leverages rich semantics from images, which are then served as additional soft supervision for training detectors. Specifically, we add a semantic branch to our detector to learn these soft semantics and enhance feature representations for long-tailed object detection. The semantic branch is only used for training and is removed during inference. RichSem achieves consistent improvements on both overall and rare-category of LVIS under different backbones and detectors. Our method achieves state-of-the-art performance without requiring complex training and testing procedures. Moreover, we show the effectiveness of our method on other long-tailed datasets with additional experiments. Code is available at \url{https://github.com/MengLcool/RichSem}.

Ziyu Jiang, Yinpeng Chen, Mengchen Liu et al.

Recently, both Contrastive Learning (CL) and Mask Image Modeling (MIM) demonstrate that self-supervision is powerful to learn good representations. However, naively combining them is far from success. In this paper, we start by making the empirical observation that a naive joint optimization of CL and MIM losses leads to conflicting gradient directions - more severe as the layers go deeper. This motivates us to shift the paradigm from combining loss at the end, to choosing the proper learning method per network layer. Inspired by experimental observations, we find that MIM and CL are suitable to lower and higher layers, respectively. We hence propose to combine them in a surprisingly simple, "sequential cascade" fashion: early layers are first trained under one MIM loss, on top of which latter layers continue to be trained under another CL loss. The proposed Layer Grafted Pre-training learns good visual representations that demonstrate superior label efficiency in downstream applications, in particular yielding strong few-shot performance besides linear evaluation. For instance, on ImageNet-1k, Layer Grafted Pre-training yields 65.5% Top-1 accuracy in terms of 1% few-shot learning with ViT-B/16, which improves MIM and CL baselines by 14.4% and 2.1% with no bells and whistles. The code is available at https://github.com/VITA-Group/layerGraftedPretraining_ICLR23.git.

Cheng-Fu Yang, Yen-Chun Chen, Jianwei Yang et al.

End-to-end Transformers have demonstrated an impressive success rate for Embodied Instruction Following when the environment has been seen in training. However, they tend to struggle when deployed in an unseen environment. This lack of generalizability is due to the agent's insensitivity to subtle changes in natural language instructions. To mitigate this issue, we propose explicitly aligning the agent's hidden states with the instructions via contrastive learning. Nevertheless, the semantic gap between high-level language instructions and the agent's low-level action space remains an obstacle. Therefore, we further introduce a novel concept of meta-actions to bridge the gap. Meta-actions are ubiquitous action patterns that can be parsed from the original action sequence. These patterns represent higher-level semantics that are intuitively aligned closer to the instructions. When meta-actions are applied as additional training signals, the agent generalizes better to unseen environments. Compared to a strong multi-modal Transformer baseline, we achieve a significant 4.5% absolute gain in success rate in unseen environments of ALFRED Embodied Instruction Following. Additional analysis shows that the contrastive objective and meta-actions are complementary in achieving the best results, and the resulting agent better aligns its states with corresponding instructions, making it more suitable for real-world embodied agents. The code is available at: https://github.com/joeyy5588/LACMA.

Munan Ning, Bin Zhu, Yujia Xie et al.

Video-based large language models (Video-LLMs) have been recently introduced, targeting both fundamental improvements in perception and comprehension, and a diverse range of user inquiries. In pursuit of the ultimate goal of achieving artificial general intelligence, a truly intelligent Video-LLM model should not only see and understand the surroundings, but also possess human-level commonsense, and make well-informed decisions for the users. To guide the development of such a model, the establishment of a robust and comprehensive evaluation system becomes crucial. To this end, this paper proposes \textit{Video-Bench}, a new comprehensive benchmark along with a toolkit specifically designed for evaluating Video-LLMs. The benchmark comprises 10 meticulously crafted tasks, evaluating the capabilities of Video-LLMs across three distinct levels: Video-exclusive Understanding, Prior Knowledge-based Question-Answering, and Comprehension and Decision-making. In addition, we introduce an automatic toolkit tailored to process model outputs for various tasks, facilitating the calculation of metrics and generating convenient final scores. We evaluate 8 representative Video-LLMs using \textit{Video-Bench}. The findings reveal that current Video-LLMs still fall considerably short of achieving human-like comprehension and analysis of real-world videos, offering valuable insights for future research directions. The benchmark and toolkit are available at: \url{https://github.com/PKU-YuanGroup/Video-Bench}.

Junke Wang, Dongdong Chen, Zuxuan Wu et al.

This paper presents OmniVL, a new foundation model to support both image-language and video-language tasks using one universal architecture. It adopts a unified transformer-based visual encoder for both image and video inputs, and thus can perform joint image-language and video-language pretraining. We demonstrate, for the first time, such a paradigm benefits both image and video tasks, as opposed to the conventional one-directional transfer (e.g., use image-language to help video-language). To this end, we propose a decoupled joint pretraining of image-language and video-language to effectively decompose the vision-language modeling into spatial and temporal dimensions and obtain performance boost on both image and video tasks. Moreover, we introduce a novel unified vision-language contrastive (UniVLC) loss to leverage image-text, video-text, image-label (e.g., image classification), video-label (e.g., video action recognition) data together, so that both supervised and noisily supervised pretraining data are utilized as much as possible. Without incurring extra task-specific adaptors, OmniVL can simultaneously support visual only tasks (e.g., image classification, video action recognition), cross-modal alignment tasks (e.g., image/video-text retrieval), and multi-modal understanding and generation tasks (e.g., image/video question answering, captioning). We evaluate OmniVL on a wide range of downstream tasks and achieve state-of-the-art or competitive results with similar model size and data scale.

Bin Xiao, Haiping Wu, Weijian Xu et al.

We introduce Florence-2, a novel vision foundation model with a unified, prompt-based representation for a variety of computer vision and vision-language tasks. While existing large vision models excel in transfer learning, they struggle to perform a diversity of tasks with simple instructions, a capability that implies handling the complexity of various spatial hierarchy and semantic granularity. Florence-2 was designed to take text-prompt as task instructions and generate desirable results in text forms, whether it be captioning, object detection, grounding or segmentation. This multi-task learning setup demands large-scale, high-quality annotated data. To this end, we co-developed FLD-5B that consists of 5.4 billion comprehensive visual annotations on 126 million images, using an iterative strategy of automated image annotation and model refinement. We adopted a sequence-to-sequence structure to train Florence-2 to perform versatile and comprehensive vision tasks. Extensive evaluations on numerous tasks demonstrated Florence-2 to be a strong vision foundation model contender with unprecedented zero-shot and fine-tuning capabilities.

Jinnian Zhang, Houwen Peng, Kan Wu et al.

Vision Transformer (ViT) models have recently drawn much attention in computer vision due to their high model capability. However, ViT models suffer from huge number of parameters, restricting their applicability on devices with limited memory. To alleviate this problem, we propose MiniViT, a new compression framework, which achieves parameter reduction in vision transformers while retaining the same performance. The central idea of MiniViT is to multiplex the weights of consecutive transformer blocks. More specifically, we make the weights shared across layers, while imposing a transformation on the weights to increase diversity. Weight distillation over self-attention is also applied to transfer knowledge from large-scale ViT models to weight-multiplexed compact models. Comprehensive experiments demonstrate the efficacy of MiniViT, showing that it can reduce the size of the pre-trained Swin-B transformer by 48\%, while achieving an increase of 1.0\% in Top-1 accuracy on ImageNet. Moreover, using a single-layer of parameters, MiniViT is able to compress DeiT-B by 9.7 times from 86M to 9M parameters, without seriously compromising the performance. Finally, we verify the transferability of MiniViT by reporting its performance on downstream benchmarks. Code and models are available at here.

Weilun Wang, Jianmin Bao, Wengang Zhou et al.

We present SinDiffusion, leveraging denoising diffusion models to capture internal distribution of patches from a single natural image. SinDiffusion significantly improves the quality and diversity of generated samples compared with existing GAN-based approaches. It is based on two core designs. First, SinDiffusion is trained with a single model at a single scale instead of multiple models with progressive growing of scales which serves as the default setting in prior work. This avoids the accumulation of errors, which cause characteristic artifacts in generated results. Second, we identify that a patch-level receptive field of the diffusion network is crucial and effective for capturing the image's patch statistics, therefore we redesign the network structure of the diffusion model. Coupling these two designs enables us to generate photorealistic and diverse images from a single image. Furthermore, SinDiffusion can be applied to various applications, i.e., text-guided image generation, and image outpainting, due to the inherent capability of diffusion models. Extensive experiments on a wide range of images demonstrate the superiority of our proposed method for modeling the patch distribution.

Junke Wang, Dongdong Chen, Chong Luo et al.

Existing deep video models are limited by specific tasks, fixed input-output spaces, and poor generalization capabilities, making it difficult to deploy them in real-world scenarios. In this paper, we present our vision for multimodal and versatile video understanding and propose a prototype system, \system. Our system is built upon a tracklet-centric paradigm, which treats tracklets as the basic video unit and employs various Video Foundation Models (ViFMs) to annotate their properties e.g., appearance, motion, \etc. All the detected tracklets are stored in a database and interact with the user through a database manager. We have conducted extensive case studies on different types of in-the-wild videos, which demonstrates the effectiveness of our method in answering various video-related problems. Our project is available at https://www.wangjunke.info/ChatVideo/

Junke Wang, Dongdong Chen, Zuxuan Wu et al.

Exploring dense matching between the current frame and past frames for long-range context modeling, memory-based methods have demonstrated impressive results in video object segmentation (VOS) recently. Nevertheless, due to the lack of instance understanding ability, the above approaches are oftentimes brittle to large appearance variations or viewpoint changes resulted from the movement of objects and cameras. In this paper, we argue that instance understanding matters in VOS, and integrating it with memory-based matching can enjoy the synergy, which is intuitively sensible from the definition of VOS task, \ie, identifying and segmenting object instances within the video. Towards this goal, we present a two-branch network for VOS, where the query-based instance segmentation (IS) branch delves into the instance details of the current frame and the VOS branch performs spatial-temporal matching with the memory bank. We employ the well-learned object queries from IS branch to inject instance-specific information into the query key, with which the instance-augmented matching is further performed. In addition, we introduce a multi-path fusion block to effectively combine the memory readout with multi-scale features from the instance segmentation decoder, which incorporates high-resolution instance-aware features to produce final segmentation results. Our method achieves state-of-the-art performance on DAVIS 2016/2017 val (92.6% and 87.1%), DAVIS 2017 test-dev (82.8%), and YouTube-VOS 2018/2019 val (86.3% and 86.3%), outperforming alternative methods by clear margins.

Fuhui Zhou, Yihao Li, Ming Xu et al.

Semantic communication is envisioned as a promising technique to break through the Shannon limit. However, semantic inference and semantic error correction have not been well studied. Moreover, error correction methods of existing semantic communication frameworks are inexplicable and inflexible, which limits the achievable performance. In this paper, to tackle this issue, a knowledge graph is exploited to develop semantic communication systems. Two cognitive semantic communication frameworks are proposed for the single-user and multiple-user communication scenarios. Moreover, a simple, general, and interpretable semantic alignment algorithm for semantic information detection is proposed. Furthermore, an effective semantic correction algorithm is proposed by mining the inference rule from the knowledge graph. Additionally, the pre-trained model is fine-tuned to recover semantic information. For the multi-user cognitive semantic communication system, a message recovery algorithm is proposed to distinguish messages of different users by matching the knowledge level between the source and the destination. Extensive simulation results conducted on a public dataset demonstrate that our proposed single-user and multi-user cognitive semantic communication systems are superior to benchmark communication systems in terms of the data compression rate and communication reliability. Finally, we present realistic single-user and multi-user cognitive semantic communication systems results by building a software-defined radio prototype system.

Lemeng Wu, Mengchen Liu, Yinpeng Chen et al.

Mixture of Experts (MoE) is able to scale up vision transformers effectively. However, it requires prohibiting computation resources to train a large MoE transformer. In this paper, we propose Residual Mixture of Experts (RMoE), an efficient training pipeline for MoE vision transformers on downstream tasks, such as segmentation and detection. RMoE achieves comparable results with the upper-bound MoE training, while only introducing minor additional training cost than the lower-bound non-MoE training pipelines. The efficiency is supported by our key observation: the weights of an MoE transformer can be factored into an input-independent core and an input-dependent residual. Compared with the weight core, the weight residual can be efficiently trained with much less computation resource, e.g., finetuning on the downstream data. We show that, compared with the current MoE training pipeline, we get comparable results while saving over 30% training cost. When compared with state-of-the-art non- MoE transformers, such as Swin-T / CvT-13 / Swin-L, we get +1.1 / 0.9 / 1.0 mIoU gain on ADE20K segmentation and +1.4 / 1.6 / 0.6 AP gain on MS-COCO object detection task with less than 3% additional training cost.

Junke Wang, Zuxuan Wu, Dongdong Chen et al.

Visual Object Tracking (VOT) aims to estimate the positions of target objects in a video sequence, which is an important vision task with various real-world applications. Depending on whether the initial states of target objects are specified by provided annotations in the first frame or the categories, VOT could be classified as instance tracking (e.g., SOT and VOS) and category tracking (e.g., MOT, MOTS, and VIS) tasks. Different definitions have led to divergent solutions for these two types of tasks, resulting in redundant training expenses and parameter overhead. In this paper, combing the advantages of the best practices developed in both communities, we propose a novel tracking-with-detection paradigm, where tracking supplements appearance priors for detection and detection provides tracking with candidate bounding boxes for the association. Equipped with such a design, a unified tracking model, OmniTracker, is further presented to resolve all the tracking tasks with a fully shared network architecture, model weights, and inference pipeline, eliminating the need for task-specific architectures and reducing redundancy in model parameters. We conduct extensive experimentation on seven prominent tracking datasets of different tracking tasks, including LaSOT, TrackingNet, DAVIS16-17, MOT17, MOTS20, and YTVIS19, and demonstrate that OmniTracker achieves on-par or even better results than both task-specific and unified tracking models.

Lingchen Meng, Xiyang Dai, Yinpeng Chen et al.

Combining multiple datasets enables performance boost on many computer vision tasks. But similar trend has not been witnessed in object detection when combining multiple datasets due to two inconsistencies among detection datasets: taxonomy difference and domain gap. In this paper, we address these challenges by a new design (named Detection Hub) that is dataset-aware and category-aligned. It not only mitigates the dataset inconsistency but also provides coherent guidance for the detector to learn across multiple datasets. In particular, the dataset-aware design is achieved by learning a dataset embedding that is used to adapt object queries as well as convolutional kernels in detection heads. The categories across datasets are semantically aligned into a unified space by replacing one-hot category representations with word embedding and leveraging the semantic coherence of language embedding. Detection Hub fulfills the benefits of large data on object detection. Experiments demonstrate that joint training on multiple datasets achieves significant performance gains over training on each dataset alone. Detection Hub further achieves SoTA performance on UODB benchmark with wide variety of datasets.

Yunsheng Li, Yinpeng Chen, Xiyang Dai et al.

The complexity-precision trade-off of an object detector is a critical problem for resource constrained vision tasks. Previous works have emphasized detectors implemented with efficient backbones. The impact on this trade-off of proposal processing by the detection head is investigated in this work. It is hypothesized that improved detection efficiency requires a paradigm shift, towards the unequal processing of proposals, assigning more computation to good proposals than poor ones. This results in better utilization of available computational budget, enabling higher accuracy for the same FLOPS. We formulate this as a learning problem where the goal is to assign operators to proposals, in the detection head, so that the total computational cost is constrained and the precision is maximized. The key finding is that such matching can be learned as a function that maps each proposal embedding into a one-hot code over operators. While this function induces a complex dynamic network routing mechanism, it can be implemented by a simple MLP and learned end-to-end with off-the-shelf object detectors. This 'dynamic proposal processing' (DPP) is shown to outperform state-of-the-art end-to-end object detectors (DETR, Sparse R-CNN) by a clear margin for a given computational complexity.

Yujia Xie, Luowei Zhou, Xiyang Dai et al.

People say, "A picture is worth a thousand words". Then how can we get the rich information out of the image? We argue that by using visual clues to bridge large pretrained vision foundation models and language models, we can do so without any extra cross-modal training. Thanks to the strong zero-shot capability of foundation models, we start by constructing a rich semantic representation of the image (e.g., image tags, object attributes / locations, captions) as a structured textual prompt, called visual clues, using a vision foundation model. Based on visual clues, we use large language model to produce a series of comprehensive descriptions for the visual content, which is then verified by the vision model again to select the candidate that aligns best with the image. We evaluate the quality of generated descriptions by quantitative and qualitative measurement. The results demonstrate the effectiveness of such a structured semantic representation.

Yinpeng Chen, Xiyang Dai, Dongdong Chen et al.

This paper presents a new perspective of self-supervised learning based on extending heat equation into high dimensional feature space. In particular, we remove time dependence by steady-state condition, and extend the remaining 2D Laplacian from x--y isotropic to linear correlated. Furthermore, we simplify it by splitting x and y axes as two first-order linear differential equations. Such simplification explicitly models the spatial invariance along horizontal and vertical directions separately, supporting prediction across image blocks. This introduces a very simple masked image modeling (MIM) method, named QB-Heat. QB-Heat leaves a single block with size of quarter image unmasked and extrapolates other three masked quarters linearly. It brings MIM to CNNs without bells and whistles, and even works well for pre-training light-weight networks that are suitable for both image classification and object detection without fine-tuning. Compared with MoCo-v2 on pre-training a Mobile-Former with 5.8M parameters and 285M FLOPs, QB-Heat is on par in linear probing on ImageNet, but clearly outperforms in non-linear probing that adds a transformer block before linear classifier (65.6% vs. 52.9%). When transferring to object detection with frozen backbone, QB-Heat outperforms MoCo-v2 and supervised pre-training on ImageNet by 7.9 and 4.5 AP respectively. This work provides an insightful hypothesis on the invariance within visual representation over different shapes and textures: the linear relationship between horizontal and vertical derivatives. The code will be publicly released.

Rui Wang, Zuxuan Wu, Dongdong Chen et al.

Transformer-based models have achieved top performance on major video recognition benchmarks. Benefiting from the self-attention mechanism, these models show stronger ability of modeling long-range dependencies compared to CNN-based models. However, significant computation overheads, resulted from the quadratic complexity of self-attention on top of a tremendous number of tokens, limit the use of existing video transformers in applications with limited resources like mobile devices. In this paper, we extend Mobile-Former to Video Mobile-Former, which decouples the video architecture into a lightweight 3D-CNNs for local context modeling and a Transformer modules for global interaction modeling in a parallel fashion. To avoid significant computational cost incurred by computing self-attention between the large number of local patches in videos, we propose to use very few global tokens (e.g., 6) for a whole video in Transformers to exchange information with 3D-CNNs with a cross-attention mechanism. Through efficient global spatial-temporal modeling, Video Mobile-Former significantly improves the video recognition performance of alternative lightweight baselines, and outperforms other efficient CNN-based models at the low FLOP regime from 500M to 6G total FLOPs on various video recognition tasks. It is worth noting that Video Mobile-Former is the first Transformer-based video model which constrains the computational budget within 1G FLOPs.

Hezhen Hu, Xiaoyi Dong, Jianmin Bao et al.

Pre-training is playing an increasingly important role in learning generic feature representation for Person Re-identification (ReID). We argue that a high-quality ReID representation should have three properties, namely, multi-level awareness, occlusion robustness, and cross-region invariance. To this end, we propose a simple yet effective pre-training framework, namely PersonMAE, which involves two core designs into masked autoencoders to better serve the task of Person Re-ID. 1) PersonMAE generates two regions from the given image with RegionA as the input and \textit{RegionB} as the prediction target. RegionA is corrupted with block-wise masking to mimic common occlusion in ReID and its remaining visible parts are fed into the encoder. 2) Then PersonMAE aims to predict the whole RegionB at both pixel level and semantic feature level. It encourages its pre-trained feature representations with the three properties mentioned above. These properties make PersonMAE compatible with downstream Person ReID tasks, leading to state-of-the-art performance on four downstream ReID tasks, i.e., supervised (holistic and occluded setting), and unsupervised (UDA and USL setting). Notably, on the commonly adopted supervised setting, PersonMAE with ViT-B backbone achieves 79.8% and 69.5% mAP on the MSMT17 and OccDuke datasets, surpassing the previous state-of-the-art by a large margin of +8.0 mAP, and +5.3 mAP, respectively.

Marah Abdin, Jyoti Aneja, Hany Awadalla et al. · microsoft-research, stanford

We introduce phi-3-mini, a 3.8 billion parameter language model trained on 3.3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3.5 (e.g., phi-3-mini achieves 69% on MMLU and 8.38 on MT-bench), despite being small enough to be deployed on a phone. Our training dataset is a scaled-up version of the one used for phi-2, composed of heavily filtered publicly available web data and synthetic data. The model is also further aligned for robustness, safety, and chat format. We also provide parameter-scaling results with a 7B, 14B models trained for 4.8T tokens, called phi-3-small, phi-3-medium, both significantly more capable than phi-3-mini (e.g., respectively 75%, 78% on MMLU, and 8.7, 8.9 on MT-bench). To enhance multilingual, multimodal, and long-context capabilities, we introduce three models in the phi-3.5 series: phi-3.5-mini, phi-3.5-MoE, and phi-3.5-Vision. The phi-3.5-MoE, a 16 x 3.8B MoE model with 6.6 billion active parameters, achieves superior performance in language reasoning, math, and code tasks compared to other open-source models of similar scale, such as Llama 3.1 and the Mixtral series, and on par with Gemini-1.5-Flash and GPT-4o-mini. Meanwhile, phi-3.5-Vision, a 4.2 billion parameter model derived from phi-3.5-mini, excels in reasoning tasks and is adept at handling both single-image and text prompts, as well as multi-image and text prompts.

Wan-Cyuan Fan, Yen-Chun Chen, Mengchen Liu et al.

Recent studies customizing Multimodal Large Language Models (MLLMs) for domain-specific tasks have yielded promising results, especially in the field of scientific chart comprehension. These studies generally utilize visual instruction tuning with specialized datasets to enhance question and answer (QA) accuracy within the chart domain. However, they often neglect the fundamental discrepancy between natural image-caption pre-training data and digital chart image-QA data, particularly in the models' capacity to extract underlying numeric values from charts. This paper tackles this oversight by exploring the training processes necessary to improve MLLMs' comprehension of charts. We present three key findings: (1) Incorporating raw data values in alignment pre-training markedly improves comprehension of chart data. (2) Replacing images with their textual representation randomly during end-to-end fine-tuning transfer the language reasoning capability to chart interpretation skills. (3) Requiring the model to first extract the underlying chart data and then answer the question in the fine-tuning can further improve the accuracy. Consequently, we introduce CHOPINLLM, an MLLM tailored for in-depth chart comprehension. CHOPINLLM effectively interprets various types of charts, including unannotated ones, while maintaining robust reasoning abilities. Furthermore, we establish a new benchmark to evaluate MLLMs' understanding of different chart types across various comprehension levels. Experimental results show that CHOPINLLM exhibits strong performance in understanding both annotated and unannotated charts across a wide range of types.

Chongyan Chen, Mengchen Liu, Noel Codella et al.

Visual Question Answering (VQA) entails answering questions about images. We introduce the first VQA dataset in which all contents originate from an authentic use case. Sourced from online question answering community forums, we call it VQAonline. We characterize this dataset and how it relates to eight mainstream VQA datasets. Observing that answers in our dataset tend to be much longer (i.e., a mean of 173 words) and so incompatible with standard VQA evaluation metrics, we instead utilize popular metrics for longer text evaluation for evaluating six state-of-the-art VQA models on VQAonline and report where they struggle most. Finally, we analyze which evaluation metrics align best with human judgments. To facilitate future extensions, we publicly-share the dataset at: https://vqaonline.github.io/.

Yinpeng Chen, Dongdong Chen, Xiyang Dai et al.

In this paper, we empirically reveal an invariance over images-images share a set of one-way wave equations with latent speeds. Each image is uniquely associated with a solution to these wave equations, allowing for its reconstruction with high fidelity from an initial condition. We demonstrate it using an intuitive encoder-decoder framework where each image is encoded into its corresponding initial condition (a single vector). Subsequently, the initial condition undergoes a specialized decoder, transforming the one-way wave equations into a first-order norm + linear autoregressive process. This process propagates the initial condition along the x and y directions, generating a high-resolution feature map (up to the image resolution), followed by a few convolutional layers to reconstruct image pixels. The revealed invariance, rooted in the shared wave equations, offers a fresh perspective for comprehending images, establishing a promising avenue for further exploration.

Wenyang Zhou, Lu Yuan, Shuyu Chen et al.

3D face generation has achieved high visual quality and 3D consistency thanks to the development of neural radiance fields (NeRF). Recently, to generate and edit 3D faces with NeRF representation, some methods are proposed and achieve good results in decoupling geometry and texture. The latent codes of these generative models affect the whole face, and hence modifications to these codes cause the entire face to change. However, users usually edit a local region when editing faces and do not want other regions to be affected. Since changes to the latent code affect global generation results, these methods do not allow for fine-grained control of local facial regions. To improve local controllability in NeRF-based face editing, we propose LC-NeRF, which is composed of a Local Region Generators Module and a Spatial-Aware Fusion Module, allowing for local geometry and texture control of local facial regions. Qualitative and quantitative evaluations show that our method provides better local editing than state-of-the-art face editing methods. Our method also performs well in downstream tasks, such as text-driven facial image editing.

Yuanze Lin, Yunsheng Li, Dongdong Chen et al.

In recent years, multimodal large language models (MLLMs) have made significant strides by training on vast high-quality image-text datasets, enabling them to generally understand images well. However, the inherent difficulty in explicitly conveying fine-grained or spatially dense information in text, such as masks, poses a challenge for MLLMs, limiting their ability to answer questions requiring an understanding of detailed or localized visual elements. Drawing inspiration from the Retrieval-Augmented Generation (RAG) concept, this paper proposes a new visual prompt approach to integrate fine-grained external knowledge, gleaned from specialized vision models (e.g., instance segmentation/OCR models), into MLLMs. This is a promising yet underexplored direction for enhancing MLLMs' performance. Our approach diverges from concurrent works, which transform external knowledge into additional text prompts, necessitating the model to indirectly learn the correspondence between visual content and text coordinates. Instead, we propose embedding fine-grained knowledge information directly into a spatial embedding map as a visual prompt. This design can be effortlessly incorporated into various MLLMs, such as LLaVA and Mipha, considerably improving their visual understanding performance. Through rigorous experiments, we demonstrate that our method can enhance MLLM performance across nine benchmarks, amplifying their fine-grained context-aware capabilities.

Junke Wang, Dongdong Chen, Chong Luo et al.

The core of video understanding tasks, such as recognition, captioning, and tracking, is to automatically detect objects or actions in a video and analyze their temporal evolution. Despite sharing a common goal, different tasks often rely on distinct model architectures and annotation formats. In contrast, natural language processing benefits from a unified output space, i.e., text sequences, which simplifies the training of powerful foundational language models, such as GPT-3, with extensive training corpora. Inspired by this, we seek to unify the output space of video understanding tasks by using languages as labels and additionally introducing time and box tokens. In this way, a variety of video tasks could be formulated as video-grounded token generation. This enables us to address various types of video tasks, including classification (such as action recognition), captioning (covering clip captioning, video question answering, and dense video captioning), and localization tasks (such as visual object tracking) within a fully shared encoder-decoder architecture, following a generative framework. Through comprehensive experiments, we demonstrate such a simple and straightforward idea is quite effective and can achieve state-of-the-art or competitive results on seven video benchmarks, providing a novel perspective for more universal video understanding. Code is available at https://github.com/wangjk666/OmniVid.