Video-Infinity: Distributed Long Video Generation

• URL: http://arxiv.org/abs/2406.16260v1
• Date: Mon, 24 Jun 2024 01:56:12 GMT
• Title: Video-Infinity: Distributed Long Video Generation
• Authors: Zhenxiong Tan, Xingyi Yang, Songhua Liu, Xinchao Wang,
• Abstract summary: Diffusion models have recently achieved remarkable results for video generation. Our method generates videos up to 2,300 frames in approximately 5 minutes, enabling long video generation at a speed 100 times faster than the prior methods.
• Score: 73.30145218077074
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion models have recently achieved remarkable results for video generation. Despite the encouraging performances, the generated videos are typically constrained to a small number of frames, resulting in clips lasting merely a few seconds. The primary challenges in producing longer videos include the substantial memory requirements and the extended processing time required on a single GPU. A straightforward solution would be to split the workload across multiple GPUs, which, however, leads to two issues: (1) ensuring all GPUs communicate effectively to share timing and context information, and (2) modifying existing video diffusion models, which are usually trained on short sequences, to create longer videos without additional training. To tackle these, in this paper we introduce Video-Infinity, a distributed inference pipeline that enables parallel processing across multiple GPUs for long-form video generation. Specifically, we propose two coherent mechanisms: Clip parallelism and Dual-scope attention. Clip parallelism optimizes the gathering and sharing of context information across GPUs which minimizes communication overhead, while Dual-scope attention modulates the temporal self-attention to balance local and global contexts efficiently across the devices. Together, the two mechanisms join forces to distribute the workload and enable the fast generation of long videos. Under an 8 x Nvidia 6000 Ada GPU (48G) setup, our method generates videos up to 2,300 frames in approximately 5 minutes, enabling long video generation at a speed 100 times faster than the prior methods.

Related papers

• Adaptive Caching for Faster Video Generation with Diffusion Transformers [52.73348147077075]
  Diffusion Transformers (DiTs) rely on larger models and heavier attention mechanisms, resulting in slower inference speeds. We introduce a training-free method to accelerate video DiTs, termed Adaptive Caching (AdaCache) We also introduce a Motion Regularization (MoReg) scheme to utilize video information within AdaCache, controlling the compute allocation based on motion content.
  arXiv  Detail & Related papers  (2024-11-04T18:59:44Z)
• FreeLong: Training-Free Long Video Generation with SpectralBlend Temporal Attention [57.651429116402554]
  This paper investigates a straightforward and training-free approach to extend an existing short video diffusion model for consistent long video generation. We find that directly applying the short video diffusion model to generate long videos can lead to severe video quality degradation. Motivated by this, we propose a novel solution named FreeLong to balance the frequency distribution of long video features during the denoising process.
  arXiv  Detail & Related papers  (2024-07-29T11:52:07Z)
• Multi-sentence Video Grounding for Long Video Generation [46.363084926441466]
  We propose a brave and new idea of Multi-sentence Video Grounding for Long Video Generation. Our approach seamlessly extends the development in image/video editing, video morphing and personalized generation, and video grounding to the long video generation.
  arXiv  Detail & Related papers  (2024-07-18T07:05:05Z)
• RGNet: A Unified Clip Retrieval and Grounding Network for Long Videos [16.916873537450424]
  Existing methods typically operate in two stages: clip retrieval and grounding. We propose RGNet which deeply integrates clip retrieval and grounding into a single network capable of processing long videos into multiple granular levels. RGNet surpasses prior methods, showcasing state-of-the-art performance on long video temporal grounding (LVTG) datasets MAD and Ego4D.
  arXiv  Detail & Related papers  (2023-12-11T09:12:35Z)
• ControlVideo: Training-free Controllable Text-to-Video Generation [117.06302461557044]
  ControlVideo is a framework to enable natural and efficient text-to-video generation. It generates both short and long videos within several minutes using one NVIDIA 2080Ti.
  arXiv  Detail & Related papers  (2023-05-22T14:48:53Z)
• Video Generation Beyond a Single Clip [76.5306434379088]
  Video generation models can only generate video clips that are relatively short compared with the length of real videos. To generate long videos covering diverse content and multiple events, we propose to use additional guidance to control the video generation process. The proposed approach is complementary to existing efforts on video generation, which focus on generating realistic video within a fixed time window.
  arXiv  Detail & Related papers  (2023-04-15T06:17:30Z)
• NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation [157.07019458623242]
  NUWA-XL is a novel Diffusion over Diffusion architecture for eXtremely Long generation. Our approach adopts a coarse-to-fine'' process, in which the video can be generated in parallel at the same granularity. Experiments show that our model not only generates high-quality long videos with both global and local coherence, but also decreases the average inference time from 7.55min to 26s.
  arXiv  Detail & Related papers  (2023-03-22T07:10:09Z)
• Diverse Generation from a Single Video Made Possible [24.39972895902724]
  We present a fast and practical method for video generation and manipulation from a single natural video. Our method generates more realistic and higher quality results than single-video GANs.
  arXiv  Detail & Related papers  (2021-09-17T15:12:17Z)

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.