AI Paper Atlas

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

Exploiting Unlabeled Data in CNNs by Self-Supervised Learning to Rank

Xialei Liu, Joost van de Weijer, Andrew D. Bagdanov

TPAMI 2019 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervision in supervised pipelines · 190 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 multi-task ranking loss with efficient Siamese backprop (vs RankIQA+FT) (ablation)MT-RankIQA · TID2013 (entire, 80/20 split)LCC / SROCC: further improving the accuracy by 2.8% on LCC and 2.6% on SROCCImproved
add ranking pretraining then fine-tune (vs direct fine-tuning) (ablation)RankIQA+FT · TID2013 (entire, 80/20 split)LCC: we improve on the baseline by 11%Improved
use efficient Siamese sampling instead of random pairs (ablation)RankIQA+FT · TID2013 (entire, 80/20 split)LCC: 0.775 -> 0.799Improved
use efficient Siamese sampling instead of hard negative mining (ablation)MT-RankIQA · TID2013 (entire, 80/20 split)LCC: 0.810 -> 0.827Improved
add more synthetic distortions as auxiliary data (ablation)MT-RankIQA · TID2013 unseen distortionsaverage gain (unseen): +0.016 -> +0.075Improved
add ImageNet pre-training (ablation)Basic CNN · UCF_CC_50Ave MAE: 454.45 -> 328.54Improved
add multi-scale augmentation on pretrained CNN (ablation)Basic CNN · UCF_CC_50Ave MAE: 328.54 -> 308.06Improved
ranking pretrain + fine-tune instead of direct fine-tuning from ImageNet (ablation)Ranking+FT · UCF_CC_50Ave MAE: 308.06 -> 325.73Hurt
add multi-task ranking+counting with random pairs (ablation)Multi-task · UCF_CC_50Ave MAE: 325.73 -> 292.67Improved
use efficient backprop instead of random pair sampling (ablation)Multi-task · UCF_CC_50Ave MAE: 292.67 -> 279.60Improved
use Query-by-example instead of Keyword unlabeled data (ablation)Multi-task · UCF_CC_50MAE: 279.6 -> 291.5Hurt
use Query-by-example instead of Keyword unlabeled data (ablation)Multi-task · ShanghaiTechMAE: 73.6 -> 72.0Improved

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