An Empirical Study of Autoregressive Pre-training from Videos

• URL: http://arxiv.org/abs/2501.05453v1
• Date: Thu, 09 Jan 2025 18:59:58 GMT
• Title: An Empirical Study of Autoregressive Pre-training from Videos
• Authors: Jathushan Rajasegaran, Ilija Radosavovic, Rahul Ravishankar, Yossi Gandelsman, Christoph Feichtenhofer, Jitendra Malik,
• Abstract summary: We treat videos as visual tokens and train transformer models to autoregressively predict future tokens. Our models are pre-trained on a diverse dataset of videos and images comprising over 1 trillion visual tokens. Our results demonstrate that, despite minimal inductive biases, autoregressive pre-training leads to competitive performance.
• Score: 67.15356613065542
• License:
• Abstract: We empirically study autoregressive pre-training from videos. To perform our study, we construct a series of autoregressive video models, called Toto. We treat videos as sequences of visual tokens and train transformer models to autoregressively predict future tokens. Our models are pre-trained on a diverse dataset of videos and images comprising over 1 trillion visual tokens. We explore different architectural, training, and inference design choices. We evaluate the learned visual representations on a range of downstream tasks including image recognition, video classification, object tracking, and robotics. Our results demonstrate that, despite minimal inductive biases, autoregressive pre-training leads to competitive performance across all benchmarks. Finally, we find that scaling our video models results in similar scaling curves to those seen in language models, albeit with a different rate. More details at https://brjathu.github.io/toto/

Related papers

• Video In-context Learning [46.40277880351059]
  In this paper, we study video in-context learning, where the model starts from an existing video clip and generates diverse potential future sequences. To achieve this, we provide a clear definition of the task, and train an autoregressive Transformer on video datasets. We design various evaluation metrics, including both objective and subjective measures, to demonstrate the visual quality and semantic accuracy of generation results.
  arXiv  Detail & Related papers  (2024-07-10T04:27:06Z)
• Predicting Long-horizon Futures by Conditioning on Geometry and Time [49.86180975196375]
  We explore the task of generating future sensor observations conditioned on the past. We leverage the large-scale pretraining of image diffusion models which can handle multi-modality. We create a benchmark for video prediction on a diverse set of videos spanning indoor and outdoor scenes.
  arXiv  Detail & Related papers  (2024-04-17T16:56:31Z)
• Any-point Trajectory Modeling for Policy Learning [64.23861308947852]
  We introduce Any-point Trajectory Modeling (ATM) to predict future trajectories of arbitrary points within a video frame. ATM outperforms strong video pre-training baselines by 80% on average. We show effective transfer learning of manipulation skills from human videos and videos from a different robot morphology.
  arXiv  Detail & Related papers  (2023-12-28T23:34:43Z)
• Early Action Recognition with Action Prototypes [62.826125870298306]
  We propose a novel model that learns a prototypical representation of the full action for each class. We decompose the video into short clips, where a visual encoder extracts features from each clip independently. Later, a decoder aggregates together in an online fashion features from all the clips for the final class prediction.
  arXiv  Detail & Related papers  (2023-12-11T18:31:13Z)
• 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)
• Learning Video Models from Text: Zero-Shot Anticipation for Procedural Actions [30.88621433812347]
  This paper presents a hierarchical model that generalizes instructional knowledge from large-scale text-corpora and transfers the knowledge to video. Given a portion of an instructional video, our model recognizes and predicts coherent and plausible actions multiple steps into the future, all in rich natural language.
  arXiv  Detail & Related papers  (2021-06-06T15:43:39Z)
• ViViT: A Video Vision Transformer [75.74690759089529]
  We present pure-transformer based models for video classification. Our model extracts-temporal tokens from the input video, which are then encoded by a series of transformer layers. We show how we can effectively regularise the model during training and leverage pretrained image models to be able to train on comparatively small datasets.
  arXiv  Detail & Related papers  (2021-03-29T15:27:17Z)
• Self-supervised Video Representation Learning by Pace Prediction [48.029602040786685]
  This paper addresses the problem of self-supervised video representation learning from a new perspective -- by video pace prediction. It stems from the observation that human visual system is sensitive to video pace. We randomly sample training clips in different paces and ask a neural network to identify the pace for each video clip.
  arXiv  Detail & Related papers  (2020-08-13T12:40:24Z)

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.