Zero-Annotation Object Detection with Web Knowledge Transfer
Preprint
Abstract
Object detection is one of the major problems in computer vision, and has been extensively studied. Most of the existing detection works rely on labor-intensive supervision, such as ground truth bounding boxes of objects or at least image-level annotations. On the contrary, we propose an object detection method that does not require any form of human annotation on target tasks, by exploiting freely available web images. In order to facilitate effective knowledge transfer from web images, we introduce a multi-instance multi-label domain adaption learning framework with two key innovations. First of all, we propose an instance-level adversarial domain adaptation network with attention on foreground objects to transfer the object appearances from web domain to target domain. Second, to preserve the class-specific semantic structure of transferred object features, we propose a simultaneous transfer mechanism to transfer the supervision across domains through pseudo strong label generation. With our end-to-end framework that simultaneously learns a weakly supervised detector and transfers knowledge across domains, we achieved significant improvements over baseline methods on the benchmark datasets.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Exploiting Web Images for Weakly Supervised Object DetectionIEEE Transactions on Multimedia, 2018
- Shuffle-Then-Assemble: Learning Object-Agnostic Visual Relationship Features2018
- Deep Self-Taught Learning for Weakly Supervised Object Localization2017
- Automatic Discovery, Association Estimation and Learning of Semantic Attributes for a Thousand Categories2017
- Feature Pyramid Networks for Object Detection2017
- Multiple Instance Detection Network with Online Instance Classifier Refinement2017
- Weakly Supervised Object Localization with Multi-Fold Multiple Instance LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Attribute-Based Classification for Zero-Shot Visual Object CategorizationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Selective Search for Object RecognitionInternational Journal of Computer Vision, 2013
- OPTIMOL: Automatic Online Picture Collection via Incremental Model LearningInternational Journal of Computer Vision, 2009