Unmasked Teacher: Towards Training-Efficient Video Foundation Models

• URL: http://arxiv.org/abs/2303.16058v2
• Date: Mon, 11 Mar 2024 09:21:50 GMT
• Title: Unmasked Teacher: Towards Training-Efficient Video Foundation Models
• Authors: Kunchang Li, Yali Wang, Yizhuo Li, Yi Wang, Yinan He, Limin Wang, Yu Qiao
• Abstract summary: Video Foundation Models (VFMs) have received limited exploration due to high computational costs and data scarcity. This paper proposes a training-efficient method for temporal-sensitive VFMs that integrates the benefits of existing methods. Our model can handle various tasks including scene-related, temporal-related, and complex video-language understanding.
• Score: 50.19560876891811
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Video Foundation Models (VFMs) have received limited exploration due to high computational costs and data scarcity. Previous VFMs rely on Image Foundation Models (IFMs), which face challenges in transferring to the video domain. Although VideoMAE has trained a robust ViT from limited data, its low-level reconstruction poses convergence difficulties and conflicts with high-level cross-modal alignment. This paper proposes a training-efficient method for temporal-sensitive VFMs that integrates the benefits of existing methods. To increase data efficiency, we mask out most of the low-semantics video tokens, but selectively align the unmasked tokens with IFM, which serves as the UnMasked Teacher (UMT). By providing semantic guidance, our method enables faster convergence and multimodal friendliness. With a progressive pre-training framework, our model can handle various tasks including scene-related, temporal-related, and complex video-language understanding. Using only public sources for pre-training in 6 days on 32 A100 GPUs, our scratch-built ViT-L/16 achieves state-of-the-art performances on various video tasks. The code and models will be released at https://github.com/OpenGVLab/unmasked_teacher.

Related papers

• E-ViLM: Efficient Video-Language Model via Masked Video Modeling with Semantic Vector-Quantized Tokenizer [5.7254320553764]
  E-ViLM is able to learn expressive representations from Video-Language corpus and generalize well to extensive Video-Language tasks. Our model reaches $39.3$% Top-$1$ accuracy on the MSRVTT benchmark, retaining $91.4$% of the accuracy of state-of-the-art larger VL architecture.
  arXiv  Detail & Related papers  (2023-11-28T22:57:17Z)
• Harvest Video Foundation Models via Efficient Post-Pretraining [67.30842563833185]
  We propose an efficient framework to harvest video foundation models from image ones. Our method is intuitively simple by randomly dropping input video patches and masking out input text during the post-pretraining procedure. Our method achieves state-of-the-art performances, which are comparable to some heavily pretrained video foundation models.
  arXiv  Detail & Related papers  (2023-10-30T14:06:16Z)
• VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking [57.552798046137646]
  Video masked autoencoder (VideoMAE) is a scalable and general self-supervised pre-trainer for building video foundation models. We successfully train a video ViT model with a billion parameters, which achieves a new state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-03-29T14:28:41Z)
• Masked Video Distillation: Rethinking Masked Feature Modeling for Self-supervised Video Representation Learning [123.63301596019522]
  Masked video distillation (MVD) is a simple yet effective two-stage masked feature modeling framework for video representation learning. For the choice of teacher models, we observe that students taught by video teachers perform better on temporally-heavy video tasks. We design a spatial-temporal co-teaching method for MVD to leverage the advantage of different teachers.
  arXiv  Detail & Related papers  (2022-12-08T18:59:59Z)
• Multi-Modal Few-Shot Temporal Action Detection [157.96194484236483]
  Few-shot (FS) and zero-shot (ZS) learning are two different approaches for scaling temporal action detection to new classes. We introduce a new multi-modality few-shot (MMFS) TAD problem, which can be considered as a marriage of FS-TAD and ZS-TAD.
  arXiv  Detail & Related papers  (2022-11-27T18:13:05Z)
• VIOLET : End-to-End Video-Language Transformers with Masked Visual-token Modeling [88.30109041658618]
  A great challenge in video-language (VidL) modeling lies in the disconnection between fixed video representations extracted from image/video understanding models and downstream VidL data. We present VIOLET, a fully end-to-end VIdeO-LanguagE Transformer, which adopts a video transformer to explicitly model the temporal dynamics of video inputs.
  arXiv  Detail & Related papers  (2021-11-24T18:31:20Z)
• PGT: A Progressive Method for Training Models on Long Videos [45.935259079953255]
  Main-stream method is to split a raw video into clips, leading to incomplete temporal information flow. Inspired by natural language processing techniques dealing with long sentences, we propose to treat videos as serial fragments satisfying Markov property. We empirically demonstrate that it yields significant performance improvements on different models and datasets.
  arXiv  Detail & Related papers  (2021-03-21T06:15:20Z)

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.