Generalization in Metric Learning: Should the Embedding Layer be the Embedding Layer?
arXiv:1803.03310
Abstract
This work studies deep metric learning under small to medium scale data as we believe that better generalization could be a contributing factor to the improvement of previous fine-grained image retrieval methods; it should be considered when designing future techniques. In particular, we investigate using other layers in a deep metric learning system (besides the embedding layer) for feature extraction and analyze how well they perform on training data and generalize to testing data. From this study, we suggest a new regularization practice where one can add or choose a more optimal layer for feature extraction. State-of-the-art performance is demonstrated on 3 fine-grained image retrieval benchmarks: Cars-196, CUB-200-2011, and Stanford Online Product.
arXiv commentnew version for WACV
References in corpus (14)
- In Defense of the Triplet Loss for Person Re-Identification
- Focal Loss for Dense Object Detection
- Local Similarity-Aware Deep Feature Embedding
- Highway and Residual Networks learn Unrolled Iterative Estimation
- Sampling Matters in Deep Embedding Learning
- Deep Image Retrieval: Learning global representations for image search
- Significance of Softmax-based Features in Comparison to Distance Metric Learning-based Features
- No Fuss Distance Metric Learning using Proxies
- Smart Mining for Deep Metric Learning
- Attention-based Ensemble for Deep Metric Learning
- Localizing and Orienting Street Views Using Overhead Imagery
- Learning a Metric Embedding for Face Recognition using the Multibatch Method
- Hard-Aware Deeply Cascaded Embedding
- Deep Metric Learning with Angular Loss