VGDFR: Diffusion-based Video Generation with Dynamic Latent Frame Rate

• URL: http://arxiv.org/abs/2504.12259v1
• Date: Wed, 16 Apr 2025 17:09:13 GMT
• Title: VGDFR: Diffusion-based Video Generation with Dynamic Latent Frame Rate
• Authors: Zhihang Yuan, Rui Xie, Yuzhang Shang, Hanling Zhang, Siyuan Wang, Shengen Yan, Guohao Dai, Yu Wang,
• Abstract summary: VGDFR is a training-free approach for Diffusion-based Video Generation with Dynamic Latent Frame Rate.<n>We show that VGDFR can achieve a speedup up to 3x for video generation with minimal quality degradation.
• Score: 16.826081397057774
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion Transformer(DiT)-based generation models have achieved remarkable success in video generation. However, their inherent computational demands pose significant efficiency challenges. In this paper, we exploit the inherent temporal non-uniformity of real-world videos and observe that videos exhibit dynamic information density, with high-motion segments demanding greater detail preservation than static scenes. Inspired by this temporal non-uniformity, we propose VGDFR, a training-free approach for Diffusion-based Video Generation with Dynamic Latent Frame Rate. VGDFR adaptively adjusts the number of elements in latent space based on the motion frequency of the latent space content, using fewer tokens for low-frequency segments while preserving detail in high-frequency segments. Specifically, our key contributions are: (1) A dynamic frame rate scheduler for DiT video generation that adaptively assigns frame rates for video segments. (2) A novel latent-space frame merging method to align latent representations with their denoised counterparts before merging those redundant in low-resolution space. (3) A preference analysis of Rotary Positional Embeddings (RoPE) across DiT layers, informing a tailored RoPE strategy optimized for semantic and local information capture. Experiments show that VGDFR can achieve a speedup up to 3x for video generation with minimal quality degradation.

Related papers

• Exploiting Temporal State Space Sharing for Video Semantic Segmentation [53.8810901249897]
  Video semantic segmentation (VSS) plays a vital role in understanding the temporal evolution of scenes.<n>Traditional methods often segment videos frame-by-frame or in a short temporal window, leading to limited temporal context, redundant computations, and heavy memory requirements.<n>We introduce a Temporal Video State Space Sharing architecture to leverage Mamba state space models for temporal feature sharing.<n>Our model features a selective gating mechanism that efficiently propagates relevant information across video frames, eliminating the need for a memory-heavy feature pool.
  arXiv  Detail & Related papers  (2025-03-26T01:47:42Z)
• STOP: Integrated Spatial-Temporal Dynamic Prompting for Video Understanding [48.12128042470839]
  We propose an integrated Spatial-TempOral dynamic Prompting (STOP) model. It consists of two complementary modules, the intra-frame spatial prompting and inter-frame temporal prompting. STOP consistently achieves superior performance against state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-03-20T09:16:20Z)
• DLFR-VAE: Dynamic Latent Frame Rate VAE for Video Generation [16.216254819711327]
  We propose the Dynamic Latent Frame Rate VAE (DLFR-VAE), a training-free paradigm that can make use of adaptive temporal compression in latent space.<n>Our simple but effective DLFR-VAE can function as a plug-and-play module, seamlessly integrating with existing video generation models.
  arXiv  Detail & Related papers  (2025-02-17T15:22:31Z)
• Improved Video VAE for Latent Video Diffusion Model [55.818110540710215]
  Video Autoencoder (VAE) aims to compress pixel data into low-dimensional latent space, playing an important role in OpenAI's Sora. Most of existing VAEs inflate a pretrained image VAE into the 3D causal structure for temporal-spatial compression. We propose a new KTC architecture and a group causal convolution (GCConv) module to further improve video VAE (IV-VAE)
  arXiv  Detail & Related papers  (2024-11-10T12:43:38Z)
• MotionAura: Generating High-Quality and Motion Consistent Videos using Discrete Diffusion [3.7270979204213446]
  We present four key contributions to address the challenges of video processing.<n>First, we introduce the 3D Inverted Vector-Quantization Variencoenco Autocoder.<n>Second, we present MotionAura, a text-to-video generation framework.<n>Third, we propose a spectral transformer-based denoising network.<n>Fourth, we introduce a downstream task of Sketch Guided Videopainting.
  arXiv  Detail & Related papers  (2024-10-10T07:07:56Z)
• Enhancing Video-Language Representations with Structural Spatio-Temporal Alignment [130.15775113897553]
  Finsta is a fine-grained structural-temporal alignment learning method. It consistently improves the existing 13 strong-tuning video-language models.
  arXiv  Detail & Related papers  (2024-06-27T15:23:36Z)
• Decouple Content and Motion for Conditional Image-to-Video Generation [6.634105805557556]
  conditional image-to-video (cI2V) generation is to create a believable new video by beginning with the condition, i.e., one image and text. Previous cI2V generation methods conventionally perform in RGB pixel space, with limitations in modeling motion consistency and visual continuity. We propose a novel approach by disentangling the target RGB pixels into two distinct components: spatial content and temporal motions.
  arXiv  Detail & Related papers  (2023-11-24T06:08:27Z)
• Neural Residual Radiance Fields for Streamably Free-Viewpoint Videos [69.22032459870242]
  We present a novel technique, Residual Radiance Field or ReRF, as a highly compact neural representation to achieve real-time free-view rendering on long-duration dynamic scenes. We show such a strategy can handle large motions without sacrificing quality. Based on ReRF, we design a special FVV that achieves three orders of magnitudes compression rate and provides a companion ReRF player to support online streaming of long-duration FVVs of dynamic scenes.
  arXiv  Detail & Related papers  (2023-04-10T08:36:00Z)
• You Can Ground Earlier than See: An Effective and Efficient Pipeline for Temporal Sentence Grounding in Compressed Videos [56.676761067861236]
  Given an untrimmed video, temporal sentence grounding aims to locate a target moment semantically according to a sentence query. Previous respectable works have made decent success, but they only focus on high-level visual features extracted from decoded frames. We propose a new setting, compressed-domain TSG, which directly utilizes compressed videos rather than fully-decompressed frames as the visual input.
  arXiv  Detail & Related papers  (2023-03-14T12:53:27Z)
• Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution [95.26202278535543]
  A simple solution is to split it into two sub-tasks: video frame (VFI) and video super-resolution (VSR) temporalsynthesis and spatial super-resolution are intra-related in this task. We propose a one-stage space-time video super-resolution framework, which directly synthesizes an HR slow-motion video from an LFR, LR video.
  arXiv  Detail & Related papers  (2020-02-26T16:59:48Z)

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.