Loss Functions

Triplet Loss

Introduced by Dong et al. in Triplet Loss in Siamese Network for Object Tracking

The goal of Triplet loss, in the context of Siamese Networks, is to maximize the joint probability among all score-pairs i.e. the product of all probabilities. By using its negative logarithm, we can get the loss formulation as follows:

$$ L_{t}\left(\mathcal{V}_{p}, \mathcal{V}_{n}\right)=-\frac{1}{M N} \sum_{i}^{M} \sum_{j}^{N} \log \operatorname{prob}\left(v p_{i}, v n_{j}\right) $$

where the balance weight $1/MN$ is used to keep the loss with the same scale for different number of instance sets.

Source: Triplet Loss in Siamese Network for Object Tracking


Paper Code Results Date Stars


Task Papers Share
Retrieval 76 9.57%
Metric Learning 72 9.07%
Person Re-Identification 43 5.42%
Image Retrieval 31 3.90%
Face Recognition 22 2.77%
General Classification 21 2.64%
Clustering 18 2.27%
Image Classification 17 2.14%
Few-Shot Learning 16 2.02%


Component Type
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign