DeepID-Net: multi-stage and deformable deep convolutional neural networks for object detection
Wanli Ouyang, Ping Luo, Xingyu Zeng, Shi Qiu, Yonglong Tian, Hongsheng Li, Shuo Yang, Zhe Wang, Yuanjun Xiong, Chen Qian, Zhenyao Zhu, Ruohui Wang, Chen-Change Loy, Xiaogang Wang, Xiaoou Tang the Chinese University of Hong Kong wlouyang,
Abstract
In this paper, we propose multi-stage and deformable deep convolutional neural networks for object detection. This new deep learning object detection diagram 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. With the proposed multi-stage training strategy, multiple classifiers are jointly optimized to process samples at different difficulty levels. 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 modeling averaging. The proposed approach ranked #2 in ILSVRC 2014. It improves the mean averaged precision obtained by RCNN, which is the state-of-the-art of object detection, from $31\%$ to $45\%$ . Detailed component-wise analysis is also provided through extensive experimental evaluation.
中文速览
目标检测长期面临外观变化大、目标形变复杂、特征与分类器割裂等难题，研究者提出了DeepID-Net这一多阶段可变形深度卷积神经网络框架来系统解决这些问题。核心方法包括：引入可变形约束池化层（def-pooling）显式建模物体部件的几何形变、采用多阶段联合训练让不同难度的样本被不同分类器分流处理、用带物体级标注的千类数据集进行预训练以获得更适合检测任务的特征、并通过变换网络结构和训练策略获得多样化模型再按类别差异化融合。在ILSVRC 2014目标检测挑战赛上，该方法最终排名第二，将当时最先进的RCNN方法的平均精度均值从31%大幅提升至45%。这项工作的意义在于首次将特征学习、部件形变建模、上下文推理和多模型融合统一到一个端到端可优化的框架中，为深度学习驱动的通用目标检测提供了系统性的方法论示范。
原文 arXiv:1409.3505;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1409.3505v1