Fast R-CNN
arXiv:1504.08083
Abstract
This paper proposes a Fast Region-based Convolutional Network method (Fast R-CNN) for object detection. Fast R-CNN builds on previous work to efficiently classify object proposals using deep convolutional networks. Compared to previous work, Fast R-CNN employs several innovations to improve training and testing speed while also increasing detection accuracy. Fast R-CNN trains the very deep VGG16 network 9x faster than R-CNN, is 213x faster at test-time, and achieves a higher mAP on PASCAL VOC 2012. Compared to SPPnet, Fast R-CNN trains VGG16 3x faster, tests 10x faster, and is more accurate. Fast R-CNN is implemented in Python and C++ (using Caffe) and is available under the open-source MIT License at https://github.com/rbgirshick/fast-rcnn.
To appear in ICCV 2015
Cited by in corpus (19)
- What makes for effective detection proposals?
- STC: A Simple to Complex Framework for Weakly-supervised Semantic Segmentation
- Eigen-CAM: Class Activation Map using Principal Components
- Pavement Image Datasets: A New Benchmark Dataset to Classify and Densify Pavement Distresses
- Adapting Mask-RCNN for Automatic Nucleus Segmentation
- Accelerator-Aware Pruning for Convolutional Neural Networks
- Enabling Pedestrian Safety using Computer Vision Techniques: A Case Study of the 2018 Uber Inc. Self-driving Car Crash
- Identification of Binary Neutron Star Mergers in Gravitational-Wave Data Using YOLO One-Shot Object Detection
- Multiple instance learning on deep features for weakly supervised object detection with extreme domain shifts
- Automatic characterization of boulders on planetary surfaces from high-resolution satellite images
- Deep-CEE I: Fishing for Galaxy Clusters with Deep Neural Nets
- PredNet and Predictive Coding: A Critical Review
- Learning to Group and Label Fine-Grained Shape Components
- In search of an interaction in the dark sector through Gaussian Process and ANN approaches
- HRCenterNet: An Anchorless Approach to Chinese Character Segmentation in Historical Documents
- Training Invertible Neural Networks as Autoencoders
- Advanced Methods for the Optical Quality Assurance of Silicon Sensors
- Virtual Underwater Datasets for Autonomous Inspections
- Investigation of a Machine learning methodology for the SKA pulsar search pipeline