Vehicle Recognition Benchmarks

Top-1 car classification accuracy on Stanford car dataset

  
    Method
    Accuracy (%)
  
  Sighthound
  93.6%
  
  	Krause et al. 1
  	92.8%
  
  	Lin et al. 2
  	91.3%
  
  	Zhang et al. 3
  	88.4%
  
  	Xie et al. 4
  	86.3%
  
  	Gosselin et al. 5
  	82.7%

Top-1 & top-5 car classification accuracy of compCar dataset

We compared our results with popular deep networks of GoogLeNet, Overfeat and AlexNet reported in [6].

  
    Method
    Accuracy (top1)
    Accuracy (top5)
  

  Sighthound Cloud
  95.88%
  99.53%
  

  	GoogLeNet 6
  	91.2%
    98.1%
  

  	Overfeat 6
  	87.9%
  	96.9%
  

  	AlexNet 6
  	81.9%
  	94.0%
  

References
1. Krause, Jonathan, e.a.: Fine-grained recognition without part annotations. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015)
2. Lin, T.Y., RoyChowdhury, A., Maji., S.: Bilinear cnn models for fine-grained visual recognition. In: Proceedings of the IEEE International Conference on Computer Vision. (2015)
3. et al., X.Z.: Embedding label structures for fine-grained feature representation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2016)
4. Xie, Saining, e.a.: Hyper-class augmented and regularized deep learning for finegrained image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015)
5. Gosselin, P.H., Murray, N., Jegou, H., Perronnin., F.: Re-visiting the fisher vector for fine-grained classification. In: Pattern Recognition Letters. (2014)
6. Yang, Linjie, e.a.: A large-scale car dataset for fine-grained categorization and verification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015)