Avoiding over-detection: towards combined object detection and counting

Jackson, Philip T.G.; Obara, Boguslaw

Abstract

Existing object detection frameworks in the deep learning field generally over-detect objects, and use non-maximum suppression (NMS) to filter out excess detections, leaving one bounding box per object. This works well so long as the ground-truth bounding boxes do not overlap heavily, as would be the case with objects that partially occlude each other, or are packed densely together. In these cases it would be beneficial, and more elegant, to have a fully end-to-end system that outputs the correct number of objects without requiring a separate NMS stage. In this paper we discuss the challenges involved in solving this problem, and demonstrate preliminary results from a prototype system.

• Presentation Conference Type: Conference Paper (Published)
• Conference Name: 16th International Conference on Artificial Intelligence and Soft Computing (ICAISC 2017).
• Start Date: Jun 1, 2017
• End Date: Jun 15, 2017
• Acceptance Date: Feb 16, 2017
• Online Publication Date: May 27, 2017
• Publication Date: May 27, 2017
• Deposit Date: Mar 16, 2017
• Publicly Available Date: May 27, 2018
• Pages: 75-85
• Series Title: Lecture notes in computer science
• Series Number: 10245
• Series ISSN: 0302-9743,1611-3349
• Book Title: Artificial intelligence and soft computing : 16th International Conference, ICAISC 2017, Zakopane, Poland, June 11-15, 2017. Proceedings. Part I.
• ISBN: 9783319590622
• DOI: https://doi.org/10.1007/978-3-319-59063-9_7
• Public URL: https://durham-repository.worktribe.com/output/1147529