Video Token Merging for Long-form Video Understanding

• URL: http://arxiv.org/abs/2410.23782v1
• Date: Thu, 31 Oct 2024 09:55:32 GMT
• Title: Video Token Merging for Long-form Video Understanding
• Authors: Seon-Ho Lee, Jue Wang, Zhikang Zhang, David Fan, Xinyu Li,
• Abstract summary: We propose a learnable video token merging algorithm that dynamically merges tokens based on their saliency. Our approach significantly reduces memory costs by 84% and boosts throughput by approximately 6.89 times compared to baseline algorithms.
• Score: 17.59960070514554
• License:
• Abstract: As the scale of data and models for video understanding rapidly expand, handling long-form video input in transformer-based models presents a practical challenge. Rather than resorting to input sampling or token dropping, which may result in information loss, token merging shows promising results when used in collaboration with transformers. However, the application of token merging for long-form video processing is not trivial. We begin with the premise that token merging should not rely solely on the similarity of video tokens; the saliency of tokens should also be considered. To address this, we explore various video token merging strategies for long-form video classification, starting with a simple extension of image token merging, moving to region-concentrated merging, and finally proposing a learnable video token merging (VTM) algorithm that dynamically merges tokens based on their saliency. Extensive experimental results show that we achieve better or comparable performances on the LVU, COIN, and Breakfast datasets. Moreover, our approach significantly reduces memory costs by 84% and boosts throughput by approximately 6.89 times compared to baseline algorithms.

Related papers

• Importance-based Token Merging for Diffusion Models [41.94334394794811]
  Diffusion models excel at high-quality image and video generation. A powerful way to speed them up is by merging similar tokens for faster computation. We show that preserving important tokens during merging significantly improves sample quality.
  arXiv  Detail & Related papers  (2024-11-23T02:01:49Z)
• ElasticTok: Adaptive Tokenization for Image and Video [109.75935878130582]
  We introduce ElasticTok, a method that conditions on prior frames to adaptively encode a frame into a variable number of tokens. During inference, ElasticTok can dynamically allocate tokens when needed. Our evaluations on images and video demonstrate the effectiveness of our approach in efficient token usage.
  arXiv  Detail & Related papers  (2024-10-10T20:54:15Z)
• SITAR: Semi-supervised Image Transformer for Action Recognition [20.609596080624662]
  This paper addresses video action recognition in a semi-supervised setting by leveraging only a handful of labeled videos. We capitalize on the vast pool of unlabeled samples and employ contrastive learning on the encoded super images. Our method demonstrates superior performance compared to existing state-of-the-art approaches for semi-supervised action recognition.
  arXiv  Detail & Related papers  (2024-09-04T17:49:54Z)
• LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models [35.88374542519597]
  Large Multimodal Models (LMMs) have shown significant visual reasoning capabilities by connecting a visual encoder and a large language model. Recent LMMs incorporate more complex visual inputs, such as high-resolution images and videos, which further increases the number of visual tokens significantly. We propose PruMerge, a novel adaptive visual token reduction strategy that significantly reduces the number of visual tokens without compromising the performance of LMMs.
  arXiv  Detail & Related papers  (2024-03-22T17:59:52Z)
• VidToMe: Video Token Merging for Zero-Shot Video Editing [100.79999871424931]
  We propose a novel approach to enhance temporal consistency in generated videos by merging self-attention tokens across frames. Our method improves temporal coherence and reduces memory consumption in self-attention computations.
  arXiv  Detail & Related papers  (2023-12-17T09:05:56Z)
• Token Fusion: Bridging the Gap between Token Pruning and Token Merging [71.84591084401458]
  Vision Transformers (ViTs) have emerged as powerful backbones in computer vision, outperforming many traditional CNNs. computational overhead, largely attributed to the self-attention mechanism, makes deployment on resource-constrained edge devices challenging. We introduce "Token Fusion" (ToFu), a method that amalgamates the benefits of both token pruning and token merging.
  arXiv  Detail & Related papers  (2023-12-02T04:29:19Z)
• CenterCLIP: Token Clustering for Efficient Text-Video Retrieval [67.21528544724546]
  In CLIP, the essential visual tokenization process, which produces discrete visual token sequences, generates many homogeneous tokens due to the redundancy nature of consecutive frames in videos. This significantly increases computation costs and hinders the deployment of video retrieval models in web applications. In this paper, we design a multi-segment token clustering algorithm to find the most representative tokens and drop the non-essential ones.
  arXiv  Detail & Related papers  (2022-05-02T12:02:09Z)
• VIMPAC: Video Pre-Training via Masked Token Prediction and Contrastive Learning [82.09856883441044]
  Video understanding relies on perceiving the global content modeling its internal connections. We propose a block-wise strategy where we mask neighboring video tokens in both spatial and temporal domains. We also add an augmentation-free contrastive learning method to further capture global content.
  arXiv  Detail & Related papers  (2021-06-21T16:48:19Z)
• Composable Augmentation Encoding for Video Representation Learning [94.2358972764708]
  We focus on contrastive methods for self-supervised video representation learning. A common paradigm in contrastive learning is to construct positive pairs by sampling different data views for the same instance, with different data instances as negatives. We propose an 'augmentation aware' contrastive learning framework, where we explicitly provide a sequence of augmentation parameterisations. We show that our method encodes valuable information about specified spatial or temporal augmentation, and in doing so also achieve state-of-the-art performance on a number of video benchmarks.
  arXiv  Detail & Related papers  (2021-04-01T16:48:53Z)

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.