DeepProposals: Hunting objects and actions by cascading deep convolutional layers
| Autor: | Ali Diba, Marco Pedersoli, Tinne Tuytelaars, Amir Ghodrati, Luc Van Gool |
|---|---|
| Přispěvatelé: | Intelligent Sensory Information Systems (IVI, FNWI) |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
FOS: Computer and information sciences
Computer science business.industry Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Inverse Pattern recognition 02 engineering and technology 010501 environmental sciences Object (computer science) 01 natural sciences Inverse cascade Object detection Action (philosophy) Artificial Intelligence Cascade Pattern recognition (psychology) 0202 electrical engineering electronic engineering information engineering Object detector 020201 artificial intelligence & image processing Computer Vision and Pattern Recognition Artificial intelligence business Software 0105 earth and related environmental sciences |
| Zdroj: | International Journal of Computer Vision, 124(2), 115-131. Springer Netherlands |
| ISSN: | 0920-5691 |
| Popis: | In this paper, a new method for generating object and action proposals in images and videos is proposed. It builds on activations of different convolutional layers of a pretrained CNN, combining the localization accuracy of the early layers with the high informative-ness (and hence recall) of the later layers. To this end, we build an inverse cascade that, going backward from the later to the earlier convolutional layers of the CNN, selects the most promising locations and refines them in a coarse-to-fine manner. The method is efficient, because i) it re-uses the same features extracted for detection, ii) it aggregates features using integral images, and iii) it avoids a dense evaluation of the proposals thanks to the use of the inverse coarse-to-fine cascade. The method is also accurate. We show that our DeepProposals outperform most of the previously proposed object proposal and action proposal approaches and, when plugged into a CNN-based object detector, produce state-of-the-art detection performance. Comment: 15 pages |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|