RepVideo: Rethinking Cross-Layer Representation for Video Generation

• URL: http://arxiv.org/abs/2501.08994v1
• Date: Wed, 15 Jan 2025 18:20:37 GMT
• Title: RepVideo: Rethinking Cross-Layer Representation for Video Generation
• Authors: Chenyang Si, Weichen Fan, Zhengyao Lv, Ziqi Huang, Yu Qiao, Ziwei Liu,
• Abstract summary: We propose RepVideo, an enhanced representation framework for text-to-video diffusion models. By accumulating features from neighboring layers to form enriched representations, this approach captures more stable semantic information. Our experiments demonstrate that our RepVideo not only significantly enhances the ability to generate accurate spatial appearances, but also improves temporal consistency in video generation.
• Score: 53.701548524818534
• License:
• Abstract: Video generation has achieved remarkable progress with the introduction of diffusion models, which have significantly improved the quality of generated videos. However, recent research has primarily focused on scaling up model training, while offering limited insights into the direct impact of representations on the video generation process. In this paper, we initially investigate the characteristics of features in intermediate layers, finding substantial variations in attention maps across different layers. These variations lead to unstable semantic representations and contribute to cumulative differences between features, which ultimately reduce the similarity between adjacent frames and negatively affect temporal coherence. To address this, we propose RepVideo, an enhanced representation framework for text-to-video diffusion models. By accumulating features from neighboring layers to form enriched representations, this approach captures more stable semantic information. These enhanced representations are then used as inputs to the attention mechanism, thereby improving semantic expressiveness while ensuring feature consistency across adjacent frames. Extensive experiments demonstrate that our RepVideo not only significantly enhances the ability to generate accurate spatial appearances, such as capturing complex spatial relationships between multiple objects, but also improves temporal consistency in video generation.

Related papers

• DiffuEraser: A Diffusion Model for Video Inpainting [13.292164408616257]
  We introduce DiffuEraser, a video inpainting model based on stable diffusion, to fill masked regions with greater details and more coherent structures. We also expand the temporal receptive fields of both the prior model and DiffuEraser, and further enhance consistency by leveraging the temporal smoothing property of Video Diffusion Models.
  arXiv  Detail & Related papers  (2025-01-17T08:03:02Z)
• Optical-Flow Guided Prompt Optimization for Coherent Video Generation [51.430833518070145]
  We propose a framework called MotionPrompt that guides the video generation process via optical flow. We optimize learnable token embeddings during reverse sampling steps by using gradients from a trained discriminator applied to random frame pairs. This approach allows our method to generate visually coherent video sequences that closely reflect natural motion dynamics, without compromising the fidelity of the generated content.
  arXiv  Detail & Related papers  (2024-11-23T12:26:52Z)
• BroadWay: Boost Your Text-to-Video Generation Model in a Training-free Way [72.1984861448374]
  We present BroadWay, a training-free method to improve the quality of text-to-video generation without introducing additional parameters, augmenting memory or sampling time. Specifically, BroadWay is composed of two principal components: 1) Temporal Self-Guidance improves the structural plausibility and temporal consistency of generated videos by reducing the disparity between the temporal attention maps across various decoder blocks.
  arXiv  Detail & Related papers  (2024-10-08T17:56:33Z)
• FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation [85.29772293776395]
  We introduce FRESCO, intra-frame correspondence alongside inter-frame correspondence to establish a more robust spatial-temporal constraint. This enhancement ensures a more consistent transformation of semantically similar content across frames. Our approach involves an explicit update of features to achieve high spatial-temporal consistency with the input video.
  arXiv  Detail & Related papers  (2024-03-19T17:59:18Z)
• APLA: Additional Perturbation for Latent Noise with Adversarial Training Enables Consistency [9.07931905323022]
  We propose a novel text-to-video (T2V) generation network structure based on diffusion models. Our approach only necessitates a single video as input and builds upon pre-trained stable diffusion networks. We leverage a hybrid architecture of transformers and convolutions to compensate for temporal intricacies, enhancing consistency between different frames within the video.
  arXiv  Detail & Related papers  (2023-08-24T07:11:00Z)
• Swap Attention in Spatiotemporal Diffusions for Text-to-Video Generation [55.36617538438858]
  We propose a novel approach that strengthens the interaction between spatial and temporal perceptions. We curate a large-scale and open-source video dataset called HD-VG-130M.
  arXiv  Detail & Related papers  (2023-05-18T11:06:15Z)
• Leveraging Local Temporal Information for Multimodal Scene Classification [9.548744259567837]
  Video scene classification models should capture the spatial (pixel-wise) and temporal (frame-wise) characteristics of a video effectively. Transformer models with self-attention which are designed to get contextualized representations for individual tokens given a sequence of tokens, are becoming increasingly popular in many computer vision tasks. We propose a novel self-attention block that leverages both local and global temporal relationships between the video frames to obtain better contextualized representations for the individual frames.
  arXiv  Detail & Related papers  (2021-10-26T19:58:32Z)
• Efficient Modelling Across Time of Human Actions and Interactions [92.39082696657874]
  We argue that current fixed-sized-temporal kernels in 3 convolutional neural networks (CNNDs) can be improved to better deal with temporal variations in the input. We study how we can better handle between classes of actions, by enhancing their feature differences over different layers of the architecture. The proposed approaches are evaluated on several benchmark action recognition datasets and show competitive results.
  arXiv  Detail & Related papers  (2021-10-05T15:39:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.