Abstract:The global self-attention mechanism in diffusion transformers involves redundant computation due to the sparse and redundant nature of visual information, and the attention map of tokens within a spatial window shows significant similarity. To address this redundancy, we propose the Proxy Token Diffusion Transformer (PT-DiT), which employs sparse representative token attention (where the number of representative tokens is much smaller than the total number of tokens) to model global visual information efficiently. Specifically, in each transformer block, we randomly sample one token from each spatial-temporal window to serve as a proxy token for that region. The global semantics are captured through the self-attention of these proxy tokens and then injected into all latent tokens via cross-attention. Simultaneously, we introduce window and shift window attention to address the limitations in detail modeling caused by the sparse attention mechanism. Building on the well-designed PT-DiT, we further develop the Qihoo-T2X family, which includes a variety of models for T2I, T2V, and T2MV tasks. Experimental results show that PT-DiT achieves competitive performance while reducing the computational complexity in both image and video generation tasks (e.g., a 48% reduction compared to DiT and a 35% reduction compared to Pixart-alpha). Our source code is available at https://github.com/360CVGroup/Qihoo-T2X.
Abstract:Data-driven approaches for autonomous driving (AD) have been widely adopted in the past decade but are confronted with dataset bias and uninterpretability. Inspired by the knowledge-driven nature of human driving, recent approaches explore the potential of large language models (LLMs) to improve understanding and decision-making in traffic scenarios. They find that the pretrain-finetune paradigm of LLMs on downstream data with the Chain-of-Thought (CoT) reasoning process can enhance explainability and scene understanding. However, such a popular strategy proves to suffer from the notorious problems of misalignment between the crafted CoTs against the consequent decision-making, which remains untouched by previous LLM-based AD methods. To address this problem, we motivate an end-to-end decision-making model based on multimodality-augmented LLM, which simultaneously executes CoT reasoning and carries out planning results. Furthermore, we propose a reasoning-decision alignment constraint between the paired CoTs and planning results, imposing the correspondence between reasoning and decision-making. Moreover, we redesign the CoTs to enable the model to comprehend complex scenarios and enhance decision-making performance. We dub our proposed large language planners with reasoning-decision alignment as RDA-Driver. Experimental evaluations on the nuScenes and DriveLM-nuScenes benchmarks demonstrate the effectiveness of our RDA-Driver in enhancing the performance of end-to-end AD systems. Specifically, our RDA-Driver achieves state-of-the-art planning performance on the nuScenes dataset with 0.80 L2 error and 0.32 collision rate, and also achieves leading results on challenging DriveLM-nuScenes benchmarks with 0.82 L2 error and 0.38 collision rate.
Abstract:General text-to-image models bring revolutionary innovation to the fields of arts, design, and media. However, when applied to garment generation, even the state-of-the-art text-to-image models suffer from fine-grained semantic misalignment, particularly concerning the quantity, position, and interrelations of garment components. Addressing this, we propose GarmentAligner, a text-to-garment diffusion model trained with retrieval-augmented multi-level corrections. To achieve semantic alignment at the component level, we introduce an automatic component extraction pipeline to obtain spatial and quantitative information of garment components from corresponding images and captions. Subsequently, to exploit component relationships within the garment images, we construct retrieval subsets for each garment by retrieval augmentation based on component-level similarity ranking and conduct contrastive learning to enhance the model perception of components from positive and negative samples. To further enhance the alignment of components across semantic, spatial, and quantitative granularities, we propose the utilization of multi-level correction losses that leverage detailed component information. The experimental findings demonstrate that GarmentAligner achieves superior fidelity and fine-grained semantic alignment when compared to existing competitors.
Abstract:Following the advancements in text-guided image generation technology exemplified by Stable Diffusion, video generation is gaining increased attention in the academic community. However, relying solely on text guidance for video generation has serious limitations, as videos contain much richer content than images, especially in terms of motion. This information can hardly be adequately described with plain text. Fortunately, in computer vision, various visual representations can serve as additional control signals to guide generation. With the help of these signals, video generation can be controlled in finer detail, allowing for greater flexibility for different applications. Integrating various controls, however, is nontrivial. In this paper, we propose a universal framework called EasyControl. By propagating and injecting condition features through condition adapters, our method enables users to control video generation with a single condition map. With our framework, various conditions including raw pixels, depth, HED, etc., can be integrated into different Unet-based pre-trained video diffusion models at a low practical cost. We conduct comprehensive experiments on public datasets, and both quantitative and qualitative results indicate that our method outperforms state-of-the-art methods. EasyControl significantly improves various evaluation metrics across multiple validation datasets compared to previous works. Specifically, for the sketch-to-video generation task, EasyControl achieves an improvement of 152.0 on FVD and 19.9 on IS, respectively, in UCF101 compared with VideoComposer. For fidelity, our model demonstrates powerful image retention ability, resulting in high FVD and IS in UCF101 and MSR-VTT compared to other image-to-video models.
Abstract:Text-Video Retrieval (TVR) aims to align and associate relevant video content with corresponding natural language queries. Most existing TVR methods are based on large-scale pre-trained vision-language models (e.g., CLIP). However, due to the inherent plain structure of CLIP, few TVR methods explore the multi-scale representations which offer richer contextual information for a more thorough understanding. To this end, we propose MUSE, a multi-scale mamba with linear computational complexity for efficient cross-resolution modeling. Specifically, the multi-scale representations are generated by applying a feature pyramid on the last single-scale feature map. Then, we employ the Mamba structure as an efficient multi-scale learner to jointly learn scale-wise representations. Furthermore, we conduct comprehensive studies to investigate different model structures and designs. Extensive results on three popular benchmarks have validated the superiority of MUSE.
Abstract:Embodied AI is transforming how AI systems interact with the physical world, yet existing datasets are inadequate for developing versatile, general-purpose agents. These limitations include a lack of standardized formats, insufficient data diversity, and inadequate data volume. To address these issues, we introduce ARIO (All Robots In One), a new data standard that enhances existing datasets by offering a unified data format, comprehensive sensory modalities, and a combination of real-world and simulated data. ARIO aims to improve the training of embodied AI agents, increasing their robustness and adaptability across various tasks and environments. Building upon the proposed new standard, we present a large-scale unified ARIO dataset, comprising approximately 3 million episodes collected from 258 series and 321,064 tasks. The ARIO standard and dataset represent a significant step towards bridging the gaps of existing data resources. By providing a cohesive framework for data collection and representation, ARIO paves the way for the development of more powerful and versatile embodied AI agents, capable of navigating and interacting with the physical world in increasingly complex and diverse ways. The project is available on https://imaei.github.io/project_pages/ario/
Abstract:Synthesizing motion-rich and temporally consistent videos remains a challenge in artificial intelligence, especially when dealing with extended durations. Existing text-to-video (T2V) models commonly employ spatial cross-attention for text control, equivalently guiding different frame generations without frame-specific textual guidance. Thus, the model's capacity to comprehend the temporal logic conveyed in prompts and generate videos with coherent motion is restricted. To tackle this limitation, we introduce FancyVideo, an innovative video generator that improves the existing text-control mechanism with the well-designed Cross-frame Textual Guidance Module (CTGM). Specifically, CTGM incorporates the Temporal Information Injector (TII), Temporal Affinity Refiner (TAR), and Temporal Feature Booster (TFB) at the beginning, middle, and end of cross-attention, respectively, to achieve frame-specific textual guidance. Firstly, TII injects frame-specific information from latent features into text conditions, thereby obtaining cross-frame textual conditions. Then, TAR refines the correlation matrix between cross-frame textual conditions and latent features along the time dimension. Lastly, TFB boosts the temporal consistency of latent features. Extensive experiments comprising both quantitative and qualitative evaluations demonstrate the effectiveness of FancyVideo. Our approach achieves state-of-the-art T2V generation results on the EvalCrafter benchmark and facilitates the synthesis of dynamic and consistent videos. The video show results can be available at https://fancyvideo.github.io/, and we will make our code and model weights publicly available.
Abstract:Empathetic response generation is designed to comprehend the emotions of others and select the most appropriate strategies to assist them in resolving emotional challenges. Empathy can be categorized into cognitive empathy and affective empathy. The former pertains to the ability to understand and discern the emotional issues and situations of others, while the latter involves the capacity to provide comfort. To enhance one's empathetic abilities, it is essential to develop both these aspects. Therefore, we develop an innovative framework that combines retrieval augmentation and emotional support strategy integration. Our framework starts with the introduction of a comprehensive emotional palette for empathy. We then apply appraisal theory to decompose this palette and create a database of empathetic responses. This database serves as an external resource and enhances the LLM's empathy by integrating semantic retrieval mechanisms. Moreover, our framework places a strong emphasis on the proper articulation of response strategies. By incorporating emotional support strategies, we aim to enrich the model's capabilities in both cognitive and affective empathy, leading to a more nuanced and comprehensive empathetic response. Finally, we extract datasets ED and ET from the empathetic dialogue dataset \textsc{EmpatheticDialogues} and ExTES based on dialogue length. Experiments demonstrate that our framework can enhance the empathy ability of LLMs from both cognitive and affective empathy perspectives. Our code is released at https://github.com/CAS-SIAT-XinHai/APTNESS.
Abstract:Image-based 3D Virtual Try-ON (VTON) aims to sculpt the 3D human according to person and clothes images, which is data-efficient (i.e., getting rid of expensive 3D data) but challenging. Recent text-to-3D methods achieve remarkable improvement in high-fidelity 3D human generation, demonstrating its potential for 3D virtual try-on. Inspired by the impressive success of personalized diffusion models (e.g., Dreambooth and LoRA) for 2D VTON, it is straightforward to achieve 3D VTON by integrating the personalization technique into the diffusion-based text-to-3D framework. However, employing the personalized module in a pre-trained diffusion model (e.g., StableDiffusion (SD)) would degrade the model's capability for multi-view or multi-domain synthesis, which is detrimental to the geometry and texture optimization guided by Score Distillation Sampling (SDS) loss. In this work, we propose a novel customizing 3D human try-on model, named \textbf{DreamVTON}, to separately optimize the geometry and texture of the 3D human. Specifically, a personalized SD with multi-concept LoRA is proposed to provide the generative prior about the specific person and clothes, while a Densepose-guided ControlNet is exploited to guarantee consistent prior about body pose across various camera views. Besides, to avoid the inconsistent multi-view priors from the personalized SD dominating the optimization, DreamVTON introduces a template-based optimization mechanism, which employs mask templates for geometry shape learning and normal/RGB templates for geometry/texture details learning. Furthermore, for the geometry optimization phase, DreamVTON integrates a normal-style LoRA into personalized SD to enhance normal map generative prior, facilitating smooth geometry modeling.
Abstract:Virtual try-on methods based on diffusion models achieve realistic try-on effects but often replicate the backbone network as a ReferenceNet or use additional image encoders to process condition inputs, leading to high training and inference costs. In this work, we rethink the necessity of ReferenceNet and image encoders and innovate the interaction between garment and person by proposing CatVTON, a simple and efficient virtual try-on diffusion model. CatVTON facilitates the seamless transfer of in-shop or worn garments of any category to target persons by simply concatenating them in spatial dimensions as inputs. The efficiency of our model is demonstrated in three aspects: (1) Lightweight network: Only the original diffusion modules are used, without additional network modules. The text encoder and cross-attentions for text injection in the backbone are removed, reducing the parameters by 167.02M. (2) Parameter-efficient training: We identified the try-on relevant modules through experiments and achieved high-quality try-on effects by training only 49.57M parameters, approximately 5.51 percent of the backbone network's parameters. (3) Simplified inference: CatVTON eliminates all unnecessary conditions and preprocessing steps, including pose estimation, human parsing, and text input, requiring only a garment reference, target person image, and mask for the virtual try-on process. Extensive experiments demonstrate that CatVTON achieves superior qualitative and quantitative results with fewer prerequisites and trainable parameters than baseline methods. Furthermore, CatVTON shows good generalization in in-the-wild scenarios despite using open-source datasets with only 73K samples.