AI Paper Atlas

Research map / Video / Video segmentation and tracking

Rethinking Self-Supervised Correspondence Learning: A Video Frame-Level Similarity Perspective

Jiarui Xu, Xiaolong Wang

ICCV 2021 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 71 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 color augmentation (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 65.0 -> 61.9Hurt
add color augmentation (ablation)VFS ResNet-18 · OTB-100precision: 48.1 -> 57.3Improved
sample different frames for predictor and target (distant sampling) (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 62.2 -> 65.0Improved
add spatial augmentation (random crop and flip) to different-frame inputs (ablation)VFS ResNet-18 · OTB-100precision: 37.4 -> 48.1Improved
increase continuous frame interval from 0 to 32 (ablation)VFS ResNet-18 · OTB-100precision: 30.6 -> 46.9Improved
use distant sampling instead of continuous interval 32 (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 64.5 -> 65.0Improved
increase sampled frames from 2 to 8 (no color aug) (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 65.0 -> 66.7Improved
increase sampled frames from 2 to 8 (with color aug) (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 61.9 -> 62.5Mixed
add negative pairs (contrastive learning) (ablation)VFS ResNet-18 · OTB-100precision: 48.1 -> 39.0Hurt
add negative pairs (contrastive learning) (ablation)VFS ResNet-18 · DAVIS-2017J&F_m: 65.0 -> 64.7Hurt
add negative pairs (contrastive learning) (ablation)VFS ResNet-18 · ImageNet-1k linear probetop-1 accuracy: 22.0 -> 24.2Improved
scale backbone from ResNet-18 to ResNet-50 (ablation)VFS (no negatives, no color aug) · DAVIS-2017J&F_m: 65.0 -> 68.9Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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