AI Paper Atlas

Research map / Video / Video segmentation and tracking

Recurrent Video Masked Autoencoders

Daniel Zoran, Nikhil Parthasarathy, Yi Yang, Drew A Hudson, João Carreira, Andrew Zisserman

CVPR 2026 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add more source frames (1 to 4) (ablation)RVM-S · SSv2accuracy: 41.020 -> 52.34Improved
add more source frames (1 to 4) (ablation)RVM-S · ScanNet depthAbsRel: 1.596 -> 1.50Improved
add more source frames (1 to 4) (ablation)RVM-S · Kineticsaccuracy: 39.280 -> 39.72No clear effect
use RNN aggregator instead of full self-attention (ablation)RVM-S · SSv2accuracy: 49.1 -> 52.34Improved
use RNN aggregator instead of full self-attention (ablation)RVM-S · ScanNet depthAbsRel: 1.505 -> 1.50No clear effect
increase training examples 250M to 2B (ablation)RVM-S · SSv2accuracy: 46.38 -> 57.20Improved
increase training examples 250M to 2B (ablation)RVM-S · Kineticsaccuracy: 33.84 -> 47.70Improved
increase training examples 250M to 2B (ablation)RVM-S · ScanNet depthAbsRel: 1.75 -> 1.200Improved
replace image encoder with recurrent video encoder (RVM-L vs DINOv2-L)RVM-L · Perception Test point trackingAJ: 36.6 -> 77.3Improved
use RVM instead of VideoMAE for spatial correspondenceRVM-L · VIPmIoU: 18.9 -> 38.0Improved
RVM-S versus distilled DINOv2-S on normalized averageRVM-S · Average over 8 tasksnormalized avg: 84.4 -> 96.1Improved
RVM-S versus distilled DINOv2-S on KineticsRVM-S · Kineticsaccuracy: 57.1 -> 49.6Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy