Region-Based Convolutional Networks for Accurate Object Detection and Segmentation
| Autor: | Ross Girshick, Jitendra Malik, Jeff Donahue, Trevor Darrell |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
business.industry
Computer science Applied Mathematics Deep learning Feature extraction 0211 other engineering and technologies Cognitive neuroscience of visual object recognition Pattern recognition 02 engineering and technology Image segmentation Object detection Support vector machine Computational Theory and Mathematics Artificial Intelligence Feature (computer vision) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Segmentation Computer vision Computer Vision and Pattern Recognition Artificial intelligence business Software 021101 geological & geomatics engineering |
| Zdroj: | IEEE transactions on pattern analysis and machine intelligence. 38(1) |
| ISSN: | 1939-3539 |
| Popis: | Object detection performance, as measured on the canonical PASCAL VOC Challenge datasets, plateaued in the final years of the competition. The best-performing methods were complex ensemble systems that typically combined multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 50 percent relative to the previous best result on VOC 2012—achieving a mAP of 62.4 percent. Our approach combines two ideas: (1) one can apply high-capacity convolutional networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data are scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, boosts performance significantly. Since we combine region proposals with CNNs, we call the resulting model an R-CNN or Region-based Convolutional Network . Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|