Data-Efficient Language-Supervised Zero-Shot Learning With Self-Distillation

Ruizhe Cheng, Bichen Wu, Peizhao Zhang, Peter Vajda, Joseph E. Gonzalez; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2021, pp. 3119-3124


Traditional computer vision models are trained to predict a fixed set of predefined categories. Recently, natural language has been shown to be a broader and richer source of supervision that provides finer descriptions to visual concepts than supervised "gold" labels. Previous works, such as CLIP, use a simple pretraining task of predicting the pairings between images and text captions. CLIP, however, is data hungry and requires more than 400M image text pairs for training. We propose a data-efficient contrastive distillation method that uses soft labels to learn from noisy image-text pairs. Our model transfers knowledge from pretrained image and sentence encoders and achieves strong performance with only 3M image text pairs, 133x smaller than CLIP. Our method exceeds the previous SoTA of general zero-shot learning on ImageNet 21k+1k by 73% relatively with a ResNet50 image encoder and DeCLUTR text encoder. We also beat CLIP by 10.5% relatively on zero-shot evaluation on Google Open Images (19,958 classes).

Related Material

[pdf] [arXiv]
@InProceedings{Cheng_2021_CVPR, author = {Cheng, Ruizhe and Wu, Bichen and Zhang, Peizhao and Vajda, Peter and Gonzalez, Joseph E.}, title = {Data-Efficient Language-Supervised Zero-Shot Learning With Self-Distillation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2021}, pages = {3119-3124} }