Fine-Grained Species Recognition With Privileged Pooling: Better Sample Efficiency Through Supervised Attention
Open access
Date
2023-12Type
- Journal Article
Abstract
We propose a scheme for supervised image classification that uses privileged information, in the form of keypoint annotations for the training data, to learn strong models from small and/or biased training sets. Our main motivation is the recognition of animal species for ecological applications such as biodiversity modelling, which is challenging because of long-tailed species distributions due to rare species, and strong dataset biases such as repetitive scene background in camera traps. To counteract these challenges, we propose a visual attention mechanism that is supervised via keypoint annotations that highlight important object parts. This privileged information, implemented as a novel privileged pooling operation, is only required during training and helps the model to focus on regions that are discriminative. In experiments with three different animal species datasets, we show that deep networks with privileged pooling can use small training sets more efficiently and generalize better. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000642595Publication status
publishedExternal links
Journal / series
IEEE Transactions on Pattern Analysis and Machine IntelligenceVolume
Pages / Article No.
Publisher
IEEESubject
Camera trap images; fine-grained species recognition; privileged pooling; supervised attention; training set biasOrganisational unit
03886 - Schindler, Konrad / Schindler, Konrad
More
Show all metadata