AttentionNet: Aggregating Weak Directions for Accurate Object Detection
arXiv:1506.07704
Abstract
We present a novel detection method using a deep convolutional neural network (CNN), named AttentionNet. We cast an object detection problem as an iterative classification problem, which is the most suitable form of a CNN. AttentionNet provides quantized weak directions pointing a target object and the ensemble of iterative predictions from AttentionNet converges to an accurate object boundary box. Since AttentionNet is a unified network for object detection, it detects objects without any separated models from the object proposal to the post bounding-box regression. We evaluate AttentionNet by a human detection task and achieve the state-of-the-art performance of 65% (AP) on PASCAL VOC 2007/2012 with an 8-layered architecture only.
To appear in ICCV 2015
References in corpus (9)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
- Return of the Devil in the Details: Delving Deep into Convolutional Nets
- Scalable, High-Quality Object Detection
- DeepID-Net: multi-stage and deformable deep convolutional neural networks for object detection
- Analyzing the Performance of Multilayer Neural Networks for Object Recognition
- Fisher Kernel for Deep Neural Activations
- Deep Poselets for Human Detection
- Do More Dropouts in Pool5 Feature Maps for Better Object Detection
Cited by in corpus (5)
- Attentional Network for Visual Object Detection
- A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
- Window-Object Relationship Guided Representation Learning for Generic Object Detections
- FPAN: Fine-grained and Progressive Attention Localization Network for Data Retrieval
- Fine-scale Surface Normal Estimation using a Single NIR Image