Hierarchical Head Design for Object Detectors
| Autor: | Frédéric Jurie, Shivang Agarwal |
|---|---|
| Přispěvatelé: | Equipe Image - Laboratoire GREYC - UMR6072, Groupe de Recherche en Informatique, Image et Instrumentation de Caen (GREYC), Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Ingénieurs de Caen (ENSICAEN), Normandie Université (NU)-Normandie Université (NU)-Université de Caen Normandie (UNICAEN), Normandie Université (NU)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Ingénieurs de Caen (ENSICAEN), Normandie Université (NU), DGA RAPID-DRAAF, Jurie, Frederic |
| Rok vydání: | 2021 |
| Předmět: |
Computer science
Computer Vision Feature extraction Inference 02 engineering and technology 010501 environmental sciences 01 natural sciences 2D Object Detection [INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] Deep Learning Bounding overwatch 0202 electrical engineering electronic engineering information engineering Computer vision 0105 earth and related environmental sciences business.industry Deep learning Detector [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] object detection Object (computer science) Object detection Anchors Pattern recognition (psychology) 020201 artificial intelligence & image processing Artificial intelligence business |
| Zdroj: | ICPR |
| DOI: | 10.1109/icpr48806.2021.9412899 |
| Popis: | The notion of anchor plays a major role in modern detection algorithms such as the Faster-RCNN [1] or the SSD detector [2]. Anchors relate the features of the last layers of the detector with bounding boxes containing objects in images. Despite their importance, the literature on object detection has not paid real attention to them. The motivation of this paper comes from the observations that (i) each anchor learns to classify and regress candidate objects independently (ii) insufficient examples are available for each anchor in case of small-scale datasets. This paper addresses these questions by proposing a novel hierarchical head for the SSD detector. The new design has the added advantage of no extra weights, as compared to the original design at inference time, while improving detectors performance for small size training sets. Improved performance on PASCAL-VOC and state-of-the-art performance on FlickrLogos-47 validate the method. We also show when the proposed design does not give additional performance gain over the original design. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|