The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition

Current approaches for fine-grained recognition do the following: First, recruit experts to annotate a dataset of images, optionally also collecting more structured data in the form of part annotations and bounding boxes. Second, train a model utilizing this data. Toward the goal of solving fine-grained recognition, we introduce an alternative approach, leveraging free, noisy data from the web and simple, generic methods of recognition. This approach has benefits in both performance and scalability. We demonstrate its efficacy on four fine-grained datasets, greatly exceeding existing state of the art without the manual collection of even a single label, and furthermore show first results at scaling to more than 10,000 fine-grained categories. Quantitatively, we achieve top-1 accuracies of 92.3% on CUB-200-2011, 85.4% on Birdsnap, 93.4% on FGVC-Aircraft, and 80.8% on Stanford Dogs without using their annotated training sets. We compare our approach to an active learning approach for expanding fine-grained datasets.

PDF Abstract

Datasets


Introduced in the Paper:

Goldfinch L-Bird

Used in the Paper:

CUB-200-2011 FGVC-Aircraft Birdsnap

Results from the Paper


Ranked #5 on Fine-Grained Image Classification on CUB-200-2011 (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Fine-Grained Image Classification CUB-200-2011 Inception-v3 Accuracy 92.3% # 5

Methods


No methods listed for this paper. Add relevant methods here