A Multi-Angle Appearance-Based Approach for Vehicle Type and Brand Recognition Utilizing Faster Regional Convolution Neural Networks
Hongying Zhang, Xusheng Li, Huazhi Yuan, Huagang Liang, Yaru Wang, Siyan Song · 2023
Read A Multi-Angle Appearance-Based Approach for Vehicle Type and Brand Recognition Utilizing Faster Regional Convolution Neural Networks by Hongying Zhang, Xusheng Li, Huazhi Yuan, Huagang Liang, Yaru Wang, Siyan Song (2023) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-10708788.txt.
Description
Hongying Zhang and co-authors propose a method to recognize vehicle type and brand from images at many angles, in a 2023 paper. Most past work used front views only, which misses the multi pose spread of real traffic images. Their method uses a fast region based neural network to find and class vehicles.
Two backbone networks pull rich features from each image. A proposal part marks regions that may hold a vehicle, and a second part refines the box and names the type and brand. The team built a public data set with five vehicle types and forty eight brands shot from many angles. Tests on that set show the method classes vehicles with accuracy.
Catalog record
| CREATOR | Hongying Zhang, Xusheng Li, Huazhi Yuan, Huagang Liang, Yaru Wang, Siyan Song |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |