- WIDER FACE: A Face Detection Benchmark CVPR 2016
- ProNet: Learning to Propose Object-specific Boxes for Cascaded Neural Networks CVPR 2016
- Deep Residual Learning for Image Recognition CVPR 2016
- Fast and accurate
- You Only Look Once: Unified, Real-Time Object Detection
- SSD: Single Shot MultiBox Detector 2016
- Joint Training of Cascaded CNN for Face Detection
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Joint Training of Cascaded CNN for Face Detection CVPR 2016
- "The proposed CNNs consist of three stages. In the first stage, it produces candidate windows quickly through a shallow CNN. Then, it refines the windows to reject a large number of non-faces windows through a more complex CNN. Finally, it uses a more powerful CNN to refine the result and output facial landmarks positions."
- Find candidates using Proposal Netowork
- Filter them using non maximum supression
- Results are passed to a second network which rejects candidates, performs bounding box regression and again Non-Maximum supression
- Have a classification network with all coins used + 1 dummmy class (for non coin detections)
- Traffic-Sign Detection and Classification in the Wild CVPR216
- Traffic sign classification
- claim that they are better than faster RCNN
- slow apporach for large images
- fully convolutional network
- they use a network for simultanious classification, detection and segmentation
- Traffic sign classification