AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Webly Supervised Fine-Grained Recognition: Benchmark Datasets and an Approach

Zeren Sun, Yazhou Yao, Xiu-Shen Wei, Yongshun Zhang, Fumin Shen, Jianxin Wu, Jian Zhang, Heng Tao Shen

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 74 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 Peer-learning instead of Co-teachingB-CNN · WebFG-496 averageaccuracy: 73.24 -> 76.46Improved
use Peer-learning instead of DecouplingB-CNN · WebFG-496 averageaccuracy: 73.84 -> 76.46Improved
use Peer-learning instead of DecouplingB-CNN · Web-Aircraftaccuracy: 75.97 -> 74.38Hurt
use Peer-learning instead of plain fine-tuningB-CNN · WebFG-496 averageaccuracy: 66.10 -> 76.46Improved
use Peer-learning with ResNet-50ResNet-50 · WebiNat-5089accuracy: 48.23 -> 54.56Improved
add web images to training dataVGG-16 · FGVC-AircraftACA: 84.8 -> 88.4Improved
add web images to training dataVGG-16 · CUB200-2011ACA: 77.7 -> 85.7Improved
add web images to training dataVGG-16 · Stanford CarsACA: 85.6 -> 92.4Improved
use NTS-Net backbone instead of B-CNNPeer-learning · WebFG-496 averageaccuracy: 76.46 -> 78.17Improved
use OPAM backbone instead of B-CNNPeer-learning · WebFG-496 averageaccuracy: 76.46 -> 75.32Hurt
apply label smoothing 0.1VGG-16 and VGG-19 · WebFG-496accuracy: smoothing parameter 0.1No clear effect
apply label smoothing 0.1ResNet-101 · WebFG-496accuracy: smoothing parameter 0.1Improved

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