AI Paper Atlas

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

Transitive Invariance for Self-Supervised Visual Representation Learning

Xiaolong Wang, Kaiming He, Abhinav Gupta

ICCV 2017 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 176 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
use transitive graph with both tracking and context invariance instead of tracking onlyTriplet-Siamese network · VOC 2007 test (Fast R-CNN)mAP: 60.2 -> 63.2Improved
use transitive graph instead of context prediction onlyTriplet-Siamese network · VOC 2007 test (Fast R-CNN)mAP: 61.5 -> 63.2Improved
use transitive graph instead of multi-task learning of both tasksTriplet-Siamese network · VOC 2007 test (Fast R-CNN)mAP: 62.1 -> 63.2Improved
use 30-frame tracks instead of 15-frame tracks (ablation)Triplet-Siamese network · VOC 2007 test (Fast R-CNN)mAP: 61.5 -> 63.2Improved
cluster with learned still-image features instead of HOG (ablation)Triplet-Siamese network · VOC 2007 test (Fast R-CNN)mAP: 60.4 -> 63.2Improved
add self-supervised pre-training versus training from scratchFast R-CNN · VOC 2007 test (Fast R-CNN)mAP: 39.7 -> 63.2Improved
use transitive self-supervised pre-training instead of ImageNetFast R-CNN · VOC 2007 test (Fast R-CNN)mAP: 67.3 -> 63.2Hurt
use transitive self-supervised pre-training instead of context pre-trainingFaster R-CNN · VOC 2007 test (Faster R-CNN, all layers fine-tuned)mAP: 62.6 -> 65.0Improved
freeze all conv layers during fine-tuning with ours vs contextFaster R-CNN · VOC 2007 test (Faster R-CNN, conv1-conv5 frozen)mAP: 38.1 -> 43.1Improved
use multi-task alternativeFaster R-CNN · VOC 2007 test (Faster R-CNN, conv1-conv5 frozen)mAP: 43.1 -> 27.6Hurt
use self-supervised pre-training instead of ImageNetFaster R-CNN · COCO minivalAP: 24.4 -> 23.5Hurt
use self-supervised pre-training instead of shorter-schedule ImageNetFaster R-CNN · COCO minivalAP: 23.7 -> 23.5No clear effect

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