DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection

Abstract

In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN \cite{girshick2014rich}, which was the state-of-the-art, from 31\% to 50.3\% on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1\%. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provide a global view for people to understand the deep learning object detection pipeline.

CVPR15, arXiv admin note: substantial text overlap with arXiv:1409.3505

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Going Deeper with Convolutions
• OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
• Visualizing and Understanding Convolutional Networks
• Deformable Part Models are Convolutional Neural Networks
• Multi-scale Orderless Pooling of Deep Convolutional Activation Features

Cited by in corpus (15)

• T-CNN: Tubelets with Convolutional Neural Networks for Object Detection from Videos
• Deformable Convolutional Networks
• Deep Autoencoder based Energy Method for the Bending, Vibration, and Buckling Analysis of Kirchhoff Plates
• Occlusion Handling in Generic Object Detection: A Review
• Deep Convolution Networks for Compression Artifacts Reduction
• Online Multi-Object Tracking Using CNN-based Single Object Tracker with Spatial-Temporal Attention Mechanism
• A Post-processing Method for Detecting Unknown Intent of Dialogue System via Pre-trained Deep Neural Network Classifier
• Crafting GBD-Net for Object Detection
• C-RPNs: Promoting Object Detection in real world via a Cascade Structure of Region Proposal Networks
• Efficient Coarse-to-Fine Non-Local Module for the Detection of Small Objects
• Structured Feature Learning for Pose Estimation
• ViP-CNN: Visual Phrase Guided Convolutional Neural Network
• Learnable Histogram: Statistical Context Features for Deep Neural Networks
• Kill Two Birds With One Stone: Boosting Both Object Detection Accuracy and Speed With adaptive Patch-of-Interest Composition
• Quantization Mimic: Towards Very Tiny CNN for Object Detection