Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
By Ross B. Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik. 2013
29,238 citations recorded by the source corpus.
Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yoshua Bengio
- Mohit Bansal
- Jianfeng Gao
- Trevor Darrell
- Quoc V. Le
- Sergey Levine
- Dacheng Tao
- Percy Liang
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Xiaogang Wang
- Ming-Hsuan Yang
- Trevor Darrell
- Jitendra Malik
- Andrew Zisserman
- Yi Yang
Citations in the corpus
- Deep Residual Learning for Image Recognition (2015, background): “Many other nontrivial visual recognition tasks [8, 12, 7, 32, 27] have also”
- Rethinking the Inception Architecture for Computer Vision (2015, background): “Also these results of table 2 suggest, one might consider using dedicated high-cost low resolution networks for smaller objects in the R-CNN [5] context.”
- Learned in Translation: Contextualized Word Vectors (2017, methodology): “Girshick et al. [2014] show that using a pretrained CNN to extract features from region proposals improves object detection and semantic segmentation models.”
- How transferable are features in deep neural networks? (2014, background): “Although examples of successful feature transfer have been reported elsewhere in the literature (Girshick et al., 2013; Donahue et al., 2013b), to our knowledge these results have been limited to noticing that transfer from a given layer is much better than the alternative of training strictly on…”
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015, methodology): “, Selective Search [4] object detectors, RCNN [5], and Fast R-CNN [2]).”
- Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model (2015, neutral): “By employing an even deeper CNN model, such as the 16-layers VGG-Net [25], they boosted the performance another 7 points [10].”
- Fast R-CNN (2015, positive): “Yet it achieves a mAP of 66.9%, which is slightly higher than the 66.0% reported for traditional R-CNN [8], even though R-CNN uses “infinite” scales in the sense that each object proposal is warped to a canonical size.”
- Object Detection Networks on Convolutional Feature Maps (2015, methodology): “the well established fine-tuning strategy [6].”
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification (2015, background): “An important application of ImageNet-trained models is for transfer learning [7] on other recognition tasks.”
- Convolutional neural networks at constrained time cost (2014, background): “Deep CNNs not only have been continuously advancing the image classification accuracy [14, 21, 24, 1, 9, 22, 23], but also play as generic feature extractors for various recognition tasks such as object detection [6, 9], semantic segmentation [6, 8], and image retrieval [14, 19].”
- Fully convolutional networks for semantic segmentation (2014, methodology): “Transfer was first demonstrated on various visual recognition tasks [3, 38], then on detection, and on both instance and semantic segmentation in hybrid proposalclassifier models [10, 15, 13].”
- Going deeper with convolutions (2014, methodology): “One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.”
- ImageNet Large Scale Visual Recognition Challenge (2014, background): “7 The Third Research Institute of the Ministry of Public Security Jie Shao, Xiaoteng Zhang, JianYing Zhou, Jian Wang, Jian Chen, Yanfeng Shang, Wenfei Wang, Lin Mei, Chuanping Hu (Girshick et al., 2014; Manen et al., 2013; Howard, 2014)”
- Caffe: Convolutional Architecture for Fast Feature Embedding (2014, methodology): “This capability is essential for tasks such as knowledge transfer [2], object detection [3], and object retrieval [5].”
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (2014, methodology): “In processing test images, our method is 24-102× faster than the R-CNN method, while achieving better or comparable accuracy on Pascal VOC 2007.”
- Microsoft COCO: Common Objects in Context (2014, methodology): “For instance the ImageNet dataset [1], which contains an unprecedented number of images, has recently enabled breakthroughs in both object classification and detection research [5], [6], [7].”
- Intriguing properties of neural networks (2013, background): “So far, unit-level inspection methods had relatively little utility beyond confirming certain intuitions regarding the complexity of the representations learned by a deep neural network [6, 13, 7, 4].”
- U-Net: Convolutional Networks for Biomedical Image Segmentation (2015, background): “//lmb.informatik.uni-freiburg.de/people/ronneber/u-net. 1 Introduction In the last two years, deep convolutional networks have outperformed the state of the art in many visual recognition tasks, e.g. [7,3]. While convolutional networks have already existed for a long time [8], their success was limi”
- Mask R-CNN (2017, methodology): “Related Work R-CNN: The Region-based CNN (R-CNN) approach [13] to bounding-box object detection is to attend to a manageable number of candidate object regions [38, 20] and evaluate convolutional networks [25, 24] independently on each RoI.”