Abstract

SOCHOR Jakub, ŠPAŇHEL Jakub and HEROUT Adam. BoxCars: 3D Boxes for Improved Fine-Grained Vehicle Recognition. The Fourth Workshop on Fine-Grained Visual Categorization (CVPR 2017). Honolulu, HI: IEEE Computer Society, 2017.
Publication language:english
Original title:BoxCars: 3D Boxes for Improved Fine-Grained Vehicle Recognition
Pages:1-2
Book:The Fourth Workshop on Fine-Grained Visual Categorization (CVPR 2017)
Conference:Computer Vision and Pattern Recognition 2017
Place:Honolulu, HI, US
Year:2017
Publisher:IEEE Computer Society
Annotation
We are dealing with the problem of fine-grained vehicle make&model recognition and verification. Our contribution is showing that extracting additional data from the video stream - besides the vehicle image itself - and feeding it into the deep convolutional neural network boosts the recognition performance considerably. This additional information includes: 3D vehicle bounding box used for "unpacking" the vehicle image, its rasterized low-resolution shape, and information about the 3D vehicle orientation. Experiments show that adding such information decreases classification error by 26% (the accuracy is improved from 0.772 to 0.832) and boosts verification average precision by 208% (0.378 to 0.785) compared to baseline pure CNN without any input modifications. This extended abstract is based on previously published CVPR paper and extended journal version of the paper which is currently under review.

Your IPv4 address: 23.22.136.56
Switch to IPv6 connection

DNSSEC [dnssec]