AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learning

Spyros Gidaris, Andrei Bursuc, Gilles Puy, Nikos Komodakis, Matthieu Cord, Patrick Perez

CVPR 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Empirical contrastive visual learning · 75 citations (OpenAlex)

Check 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
Use a queue vocabulary of locally averaged features rather than online k-means with Sinkhorn balancing (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 39.74 -> 45.86Improved
Use locally averaged instead of globally averaged queue features with multi-scale BoW (ablation)OBoW · ImageNet 20% few-shot, 1-shotaccuracy: 41.29 -> 44.18Improved
Anneal teacher momentum from 0.99 to 1.0 rather than use constant low momentum (ablation)OBoW · ImageNet 20% few-shot, 1-shotaccuracy: 12.70 -> 42.11Improved
Use dynamic weight generation instead of fixed prediction weights (ablation)OBoW · ImageNet 20% few-shot, 1-shotaccuracy: 2.00 -> 42.11Improved
Use soft rather than hard visual-word assignments (ablation)OBoW · ImageNet 20% few-shot, 1-shotaccuracy: 38.61 -> 42.11Improved
Use a smaller single crop (160²) rather than one 224² crop (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 31.39 -> 45.86Improved
Add cutmix to one 224² crop (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 31.39 -> 39.46Improved
Use two 160² crops instead of one 160² crop (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 45.86 -> 47.64Improved
Add five 96² patches to two 160² crops (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 47.64 -> 49.64Improved
Add multi-scale BoW targets to two 160² crops plus five 96² patches (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 49.64 -> 50.89Improved
Use OBoW instead of BoWNetself-supervised ResNet18 · ImageNet 20% few-shot, 1-shotaccuracy: 33.80 -> 42.11Improved
Use OBoW full rather than OBoW vanilla (ablation)OBoW · ImageNet 20% linear evaluationaccuracy: 45.86 -> 50.89Improved

Later papers that compared with this method

3 later accepted papers report OBoW as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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