Pith. sign in

REVIEW 17 cited by

VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2203.12602 v3 pith:4SUUPXFF submitted 2022-03-23 cs.CV

VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

classification cs.CV
keywords videovideomaepre-trainingdatadatasetsextraimportantmasking
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Pre-training video transformers on extra large-scale datasets is generally required to achieve premier performance on relatively small datasets. In this paper, we show that video masked autoencoders (VideoMAE) are data-efficient learners for self-supervised video pre-training (SSVP). We are inspired by the recent ImageMAE and propose customized video tube masking with an extremely high ratio. This simple design makes video reconstruction a more challenging self-supervision task, thus encouraging extracting more effective video representations during this pre-training process. We obtain three important findings on SSVP: (1) An extremely high proportion of masking ratio (i.e., 90% to 95%) still yields favorable performance of VideoMAE. The temporally redundant video content enables a higher masking ratio than that of images. (2) VideoMAE achieves impressive results on very small datasets (i.e., around 3k-4k videos) without using any extra data. (3) VideoMAE shows that data quality is more important than data quantity for SSVP. Domain shift between pre-training and target datasets is an important issue. Notably, our VideoMAE with the vanilla ViT can achieve 87.4% on Kinetics-400, 75.4% on Something-Something V2, 91.3% on UCF101, and 62.6% on HMDB51, without using any extra data. Code is available at https://github.com/MCG-NJU/VideoMAE.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. SignMAE: Segmentation-Driven Self-Supervised Learning for Sign Language Recognition

    cs.CV 2026-05 unverdicted novelty 7.0

    SignMAE uses segmentation-driven masking in a mask-and-reconstruct self-supervised task to learn fine-grained sign representations, achieving state-of-the-art accuracy on WLASL, NMFs-CSL, and Slovo with fewer frames a...

  2. Mask World Model: Predicting What Matters for Robust Robot Policy Learning

    cs.RO 2026-04 unverdicted novelty 7.0

    Mask World Model predicts semantic mask dynamics with video diffusion and integrates it with a diffusion policy head, outperforming RGB world models on LIBERO and RLBench while showing better real-world generalization...

  3. VTI-CoT: Visual-Textual Interleaved Chain of Thought for Video Reasoning

    cs.CV 2026-06 unverdicted novelty 6.0

    VTI-CoT proposes a visual-textual interleaved chain-of-thought method for video reasoning, built via automated annotation and OCR compression, claiming SOTA performance and better training efficiency on same-scale models.

  4. SpecSem-Net: Integrating Spectral and Semantic Features for Robust AI-generated Video Detection

    cs.CV 2026-05 unverdicted novelty 6.0

    SpecSem-Net integrates Fourier-based spectral filtering with semantic-guided gated merging to detect AI-generated videos, reporting 87.25% accuracy on a new benchmark of five commercial generators and 95.59% on public...

  5. Prognostic Value of Lung Ultrasound Biomarkers for Readmission Risk in Congestive Heart Failure: A Pilot Data-Driven Analysis

    eess.SP 2026-05 unverdicted novelty 6.0

    Pilot study uses pretrained video encoder features from lung ultrasound to predict 30-day CHF readmission, finding lower-lung views and temporal differences most informative with top MLP F1 of 0.80.

  6. Beyond Independent Frames: Latent Attention Masked Autoencoders for Multi-View Echocardiography

    cs.CV 2026-04 unverdicted novelty 6.0

    LAMAE adds latent-space attention to masked autoencoders so multi-view echocardiography videos can exchange information across frames and views, yielding representations that transfer from adult to pediatric hearts an...

  7. Zero-shot World Models Are Developmentally Efficient Learners

    cs.AI 2026-04 unverdicted novelty 6.0

    A zero-shot visual world model trained on one child's experience achieves broad competence on physical understanding benchmarks while matching developmental behavioral patterns.

  8. Interpreting Video Representations with Spatio-Temporal Sparse Autoencoders

    cs.CV 2026-04 conditional novelty 6.0

    Spatio-temporal contrastive SAEs recover temporal coherence lost by hard TopK, improve action probes by +3.9% and retrieval by up to 2.8× R@1, and expose a monosemanticity metric artifact.

  9. LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment

    cs.CV 2023-10 unverdicted novelty 6.0

    LanguageBind aligns video, infrared, depth, and audio to a frozen language encoder via contrastive learning on the new VIDAL-10M dataset, extending video-language pretraining to N modalities.

  10. Do Video Foundation Models Understand Intuitive Physics? A Layerwise Probing Analysis

    cs.CV 2026-06 unverdicted novelty 5.0

    Video foundation models encode intuitive physics knowledge that is strongest in V-JEPA at intermediate-to-late layers and depends on pretraining type and probe design.

  11. BioVid: Autoregressive Video Generation with Biological Behavior Semantic Comprehension

    cs.CV 2026-06 unverdicted novelty 5.0

    BioVid is a data-driven autoregressive model using 2D-encode/3D-decode tokenization and causal Transformer with EOS termination that reproduces real action duration distributions (W1 distance 1.24 frames) on NTU RGB+D...

  12. EVA-Net: Subject-Independent EEG Motor Decoding with Video-Derived Motor Priors

    cs.AI 2026-06 unverdicted novelty 5.0

    EVA-Net improves subject-independent EEG motor decoding by using video action priors via cross-modal contrastive alignment and knowledge distillation, reporting an 8.66% LOSO accuracy gain on EEGMMI.

  13. EVA-Net: Subject-Independent EEG Motor Decoding with Video-Derived Motor Priors

    cs.AI 2026-06 conditional novelty 5.0

    Aligning EEG with action-video motor priors and distilling them into an EEG-only classifier yields large subject-independent accuracy gains, including +8.66% LOSO on EEGMMI.

  14. FAST-ME: Foundation-aware Adaptive Stopping for Motion Estimation for Efficient IoT Video Analysis

    cs.CV 2026-05 unverdicted novelty 5.0

    A hybrid motion estimation framework combines optimal stopping theory with foundation model semantic scores to reduce computation while maintaining accuracy and semantic coverage in video analysis.

  15. Insights from Visual Cognition: Understanding Human Action Dynamics with Overall Glance and Refined Gaze Transformer

    cs.CV 2026-04 unverdicted novelty 5.0

    The OG-ReG Transformer achieves state-of-the-art results on Kinetics-400, Something-Something v2, and Diving-48 by combining global glance and local gaze processing paths.

  16. Empirical Evaluation of Multi-Modal Touch Detection in Over-the-Shoulder Video Surveillance

    cs.CV 2026-06 accept novelty 3.0

    An empirical evaluation of a multi-modal touch detector using MediaPipe, HSV skin filtering, motion differencing, and Canny edges finds low F1 scores on staged video and excessive false positives on real videos, concl...

  17. Deep Learning-Based Sign Language Recognition from Videos and Cross-Lingual Translation to Indian Vernaculars

    cs.AI 2026-06 conditional novelty 3.0

    Fine-tuned VideoMAE reaches 78% validation accuracy on 13 Indian sign language classes from 197 clips, followed by NLLB-200 translation to Hindi, Telugu, and Bengali.