Collaborative Learning for Weakly Supervised Object Detection

By Jiajie Wang, Jiangchao Yao, Ya Zhang, Rui Zhang. 2018

50 citations recorded by the source corpus.

Weakly supervised object detection has recently received much attention, since it only requires image-level labels instead of the bounding-box labels consumed in strongly supervised learning. Nevertheless, the save in labeling expense is usually at the cost of model accuracy.In this paper, we propose a simple but effective weakly supervised collaborative learning framework to resolve this problem, which trains a weakly supervised learner and a strongly supervised learner jointly by enforcing partial feature sharing and prediction consistency. For object detection, taking WSDDN-like architecture as weakly supervised detector sub-network and Faster-RCNN-like architecture as strongly supervised detector sub-network, we propose an end-to-end Weakly Supervised Collaborative Detection Network. As there is no strong supervision available to train the Faster-RCNN-like sub-network, a new prediction consistency loss is defined to enforce consistency of predictions between the two sub-networks as well as within the Faster-RCNN-like sub-networks. At the same time, the two detectors are designed to partially share features to further guarantee the model consistency at perceptual level. Extensive experiments on PASCAL VOC 2007 and 2012 data sets have demonstrated the effectiveness of the proposed framework.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Anelia Angelova
2. Weicheng Kuo
3. In-So Kweon
4. Dahun Kim
5. Fang Wan
6. Tsung-Yi Lin
7. Nojun Kwak
8. C. V. Jawahar

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Dacheng Tao
3. Jitendra Malik
4. Andrew Zisserman
5. Trevor Darrell
6. Cordelia Schmid
7. Xiaogang Wang
8. Ming-Hsuan Yang

In Media Technology

1. Aditya Arun
2. Anelia Angelova
3. C. V. Jawahar
4. Fang Wan
5. In-So Kweon
6. Dahun Kim
7. Nojun Kwak
8. Tsung-Yi Lin

Citations in the corpus

• Learning Open-World Object Proposals without Learning to Classify (2021, background): “Weakly-supervised approaches [26], [27], [28] aim to utilize abundant image-level labels and leverage class-agnostic box proposals to build detectors.”
• Weakly Supervised Object Localization and Detection: A Survey (2021, unknown): “The approaches of this category usually train a network to mine the initial object regions [47], [92], [172] and another network for the detection task under the MIL framework [92], [145], [159], [205].”
• C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object Detection (2019, methodology): “[37] introduce a collaborative leaning framework which combine WSDDN detector and Faster RCNN in one end-to-end network, and use feature sharing to improve WSDNN at the same time.”
• Consistency-based Semi-supervised Learning for Object detection (2019, methodology): “Weakly supervised object detection methods [4, 5, 6, 7] are required to learn only with image-level labeled data, as shown in Figure 1 (b).”
• Dissimilarity Coefficient Based Weakly Supervised Object Detection (2018, background): “[37] train a weakly and strongly supervised model jointly, in a collaborative manner.”