My Object Detection Playground

Reference
https://github.com/hoya012/deep_learning_object_detection
https://github.com/amusi/awesome-object-detection
Deep Learning for Generic Object Detection: A Survey

paper

• Using Selective Search to extract category-independent region proposals.
• Using CNN to extract a fixed-length feature vector from each region.
• Using Linear SVMs to do classification.
• Using Fine Tuning to do Bounding Box Regression.

paper

• Using selective search to extract region proposals.
• Using CNN and max pooling layers to produce a conv feature map.
• Using Region of Interests(RoI) poolin to extract a fixed-length feature vector(6*6) from the feature map.
• Using Softmax to do classification.
  Using Fine Tuning to do Bounding Box Regression.

paper
code

• Using Region Proposal Networks(RPNs) to extract region proposals.
• Using CNN and max pooling layers to produce a conv feature map.
• Using RoI pooling to extract a fixed-length feature vector(6*6) from the feature map.
• Using Softmax to do classification.
  Using Fine Tuning to do Bounding Box Regression.

paper

paper

paper
code

paper