Hard negative classes for multiple object detection

We propose an efficient method to train multiple object detectors simultaneously using a large scale image dataset. The one-vs-all approach that optimizes the boundary between positive samples from a target class and negative samples from the others has been the most standard approach for object detection. However, because this approach trains each object detector independently, the scores are not balanced between object classes. The proposed method combines ideas derived from both detection and classification in order to balance the scores across all object classes. We optimized the boundary between target classes and their “hard negative” samples, just as in detection, while simultaneously balancing the detector scores across object classes, as done in multi-class classification. We evaluated the performances on multi-class object detection using a subset of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2011 dataset and showed our method outperformed a de facto standard method.

Authors

• Asako Kanezaki
• Sho Inaba
• Yoshitaka Ushiku
• Yuya Yamashita
• Hiroshi Muraoka
• Yasuo Kuniyoshi
• Tatsuya Harada

Keywords

• Object detection
• Support vector machines
• Detectors
• Training
• Vectors
• Standards
• Optimization
• Multiple Objects
• Multiple Object Detection
• Positive Samples
• Negative Samples
• Multi-label
• Classification Of Samples
• Object Classification
• Target Class
• Multiple Detection
• Large-scale Image Datasets
• Aspect Ratio
• Classification Task
• Class Labels
• Bounding Box
• Weight Vector
• Sample Phase
• Target Object
• Negative Images
• Object Detection Task
• Conditional Random Field
• Multiple Kernel Learning
• Non-maximum Suppression
• Filter Components
• Background Class
• Object Position
• Object Detection Framework
• Non-confrontational
• Conventional Objective