AI Paper Atlas

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

The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition

Jonathan Krause, Benjamin Sapp, Andrew Howard, Howard Zhou, Alexander Toshev, Tom Duerig, James Philbin, Li Fei-Fei

ECCV 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 317 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
train on raw web search images instead of the CUB ground truth training setInception-v3 · CUB-200-2011top-1 accuracy (%): 84.4 -> 87.7Improved
add cross-category filtering of images appearing in multiple categories (CUB) (ablation)Web training · CUB-200-2011top-1 accuracy (%): 87.7 -> 89.0Improved
add filtering (Birdsnap) (ablation)Web training · Birdsnaptop-1 accuracy (%): 76.1 -> 78.2Improved
add filtering (Stanford Dogs) (ablation)Web training · Stanford Dogstop-1 accuracy (%): 78.5 -> 78.4No clear effect
pre-train on the larger L-Bird (10,982 species) before fine-tuning on CUB categories (ablation)Web (filtered) · CUB-200-2011top-1 accuracy (%): 89.0 -> 91.9Improved
pre-train on L-Bird before fine-tuning on Birdsnap categories (ablation)Web (filtered) · Birdsnaptop-1 accuracy (%): 78.2 -> 82.8Improved
pre-train on larger L-Aircraft (409 types) (ablation)Web (filtered) · FGVC-Aircrafttop-1 accuracy (%): 91.1 -> 90.9No clear effect
fine-tune additionally on ground truth FGVC training set (ablation)L-Aircraft · FGVC-Aircrafttop-1 accuracy (%): 90.9 -> 94.5Improved
fine-tune additionally on Stanford Dogs ground truth training set (ablation)L-Dog · Stanford Dogstop-1 accuracy (%): 78.4 -> 84.0Improved
use 144 test-time crops instead of single crop (CUB, L-Bird) (ablation)L-Bird · CUB-200-2011top-1 accuracy (%): 91.9 -> 92.3Improved
add second round of active learning (from scratch) (ablation)Active learning · Stanford Dogs (active learning comparison)top-1 accuracy (%): 65.8 -> 74.0Improved
subsample active learning data to size of filtered web data (ablation)A.L., one round (ft) · Stanford Dogs (active learning comparison)top-1 accuracy (%): 78.8 vs 78.4No 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