Compositional Embeddings for Multi-Label One-Shot Learning

Zeqian Li, Michael Mozer, Jacob Whitehill; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2021, pp. 296-304


We present a compositional embedding framework that infers not just a single class per input image, but a set of classes, in the setting of one-shot learning. Specifically, we propose and evaluate several novel models consisting of (1) an embedding function f trained jointly with a "composition" function g that computes set union operations between the classes encoded in two embedding vectors; and (2) embedding f trained jointly with a "query" function h that computes whether the classes encoded in one embedding subsume the classes encoded in another embedding. In contrast to prior work, these models must both perceive the classes associated with the input examples and encode the relationships between different class label sets, and they are trained using only weak one-shot supervision consisting of the label-set relationships among training examples. Experiments on the OmniGlot, Open Images, and COCO datasets show that the proposed compositional embedding models outperform existing embedding methods. Our compositional embedding models have applications to multi-label object recognition for both one-shot and supervised learning.

Related Material

[pdf] [supp] [arXiv]
@InProceedings{Li_2021_WACV, author = {Li, Zeqian and Mozer, Michael and Whitehill, Jacob}, title = {Compositional Embeddings for Multi-Label One-Shot Learning}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2021}, pages = {296-304} }