Vehicle Attribute Recognition by Appearance: Computer Vision Methods for Vehicle Type, Make and Model Classification
arXiv:2006.16400 · doi:10.1007/s11265-020-01567-6
Abstract
This paper studies vehicle attribute recognition by appearance. In the literature, image-based target recognition has been extensively investigated in many use cases, such as facial recognition, but less so in the field of vehicle attribute recognition. We survey a number of algorithms that identify vehicle properties ranging from coarse-grained level (vehicle type) to fine-grained level (vehicle make and model). Moreover, we discuss two alternative approaches for these tasks, including straightforward classification and a more flexible metric learning method. Furthermore, we design a simulated real-world scenario for vehicle attribute recognition and present an experimental comparison of the two approaches.
Published in Journal of Signal Processing Systems
References in corpus (14)
- FaceNet: A Unified Embedding for Face Recognition and Clustering
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- An Overview of Multi-Task Learning in Deep Neural Networks
- OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
- A Survey of Model Compression and Acceleration for Deep Neural Networks
- Generalizing from a Few Examples: A Survey on Few-Shot Learning
- Random Erasing Data Augmentation
- On the Variance of the Adaptive Learning Rate and Beyond
- Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
- BoxCars: Improving Fine-Grained Recognition of Vehicles using 3-D Bounding Boxes in Traffic Surveillance
- Random Feature Maps for Dot Product Kernels
- Car Type Recognition with Deep Neural Networks
- Geometry-constrained Car Recognition Using a 3D Perspective Network