Class-specific object proposals re-ranking for object detection in automatic driving
Class-specific object proposals re-ranking for object detection in automatic driving
复制标题
自动驾驶中目标检测的特定类别目标建议重新排序
DOI：
10.1016/j.neucom.2017.02.068
复制
发表时间：
2017
期刊：
影响因子：
6
通讯作者：
Li Shaozi
中科院分区：
文献类型：
--
作者：
Zhong Zhun;Lei Mingyi;Cao Donglin;Fan Jianping;Li Shaozi
目标提案生成是目标检测中的重要步骤，获得高质量的提案可以有效提高检测性能。在本文中，我们提出了一种语义的、特定于类的方法来重新排序对象提案，即使提案较少，也可以持续提高召回性能。具体来说，我们首先为每个提案提取特征，包括语义分割、立体信息、上下文信息、基于 CNN 的对象性和低级提示，然后使用结构化 SVM 学习的特定于类的权重对它们进行评分。该模型的优点有两个：1）它可以很容易地合并到现有的生成器中，计算成本很少；2）即使使用更少的建议，它也可以在严格的关键条件下实现高召回率。对 KITTI 基准的实验评估表明，我们的方法显着提高了现有流行生成器的召回性能。此外，在目标检测实验中，即使有 1500 个提案，我们的方法仍然比具有 5000 个提案的基线具有更高的平均精度（AP）。
Object proposal generation is an important step in object detection, obtaining high-quality proposals can effectively improve the performance of detection. In this paper, we propose a semantic, class-specific approach to re-rank object proposals, which can consistently improve the recall performance even with fewer proposals. Specifically, we first extract features for each proposal including semantic segmentation, stereo information, contextual information, CNN-based objectness and low-level cue, and then score them using class-specific weights learned by Structured SVM. The advantages of the proposed model are two-fold: 1) it can be easily merged to existing generators with few computational costs, and 2) it can achieve high recall rate under strict critical even using fewer proposals. Experimental evaluation on the KITTI benchmark demonstrates that our approach significantly improves existing popular generators on recall performance. Moreover, in the experiment conducted for object detection, even with 1500 proposals, our approach can still have higher average precision (AP) than baselines with 5000 proposals.