CLIP the Gap: A Single Domain Generalization Approach for Object Detection
Authors:
Top Cited Papers
- 1 June 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3219-3229
Abstract
Single Domain Generalization (SDG) tackles the problem of training a model on a single source domain so that it generalizes to any unseen target domain. While this has been well studied for image classification, the literature on SDG object detection remains almost non-existent. To address the challenges of simultaneously learning robust object localization and representation, we propose to leverage a pre-trained vision-language model to introduce semantic domain concepts via textual prompts. We achieve this via a semantic augmentation strategy acting on the features extracted by the detector backbone, as well as a text-based classification loss. Our experiments evidence the benefits of our approach, outperforming by 10% the only existing SDG object detection method, Single-DGOD [52], on their own diverse weather-driving benchmark.Keywords
All Related Versions
Funding Information
- Swiss National Science Foundation
This publication has 37 references indexed in Scilit:
- Towards Universal Object Detection by Domain Attention2019
- Domain Adaptive Faster R-CNN for Object Detection in the Wild2018
- Finding Beans in Burgers: Deep Semantic-Visual Embedding with Localization2018
- Domain Generalization with Adversarial Feature Learning2018
- Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation2018
- Semantic Foggy Scene Understanding with Synthetic DataInternational Journal of Computer Vision, 2018
- Mask R-CNN2017
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Deep Residual Learning for Image Recognition2016
- Glove: Global Vectors for Word Representation2014