论文标题
NOH-NMS：通过附近物体幻觉改善行人检测
NOH-NMS: Improving Pedestrian Detection by Nearby Objects Hallucination
论文作者
论文摘要
贪婪的NM天生就会引起困境，较低的NMS阈值可能会导致召回率较低，而较高的阈值引入了更多的假阳性。由于实例密度更加强烈，此问题在行人检测中更为严重。但是，以前关于NMS的作品并没有考虑或模糊地考虑附近行人存在的因素。因此，我们提出了附近的对象幻影仪（NOH），该物体用高斯分布和NOH-NMS一起查明附近的每个提案附近的对象，从而动态地减轻了可能包含可能具有很高可能性的其他物体的抑制。与贪婪的nms相比，我们的方法（作为最先进的方法）提高了$ 3.9 \％$ ap，$ 5.1 \％$召回和$ 0.8 \％$ $ $ $ $ \ text {mr}^{ - 2} $ on craverhuman上 分别。
Greedy-NMS inherently raises a dilemma, where a lower NMS threshold will potentially lead to a lower recall rate and a higher threshold introduces more false positives. This problem is more severe in pedestrian detection because the instance density varies more intensively. However, previous works on NMS don't consider or vaguely consider the factor of the existent of nearby pedestrians. Thus, we propose Nearby Objects Hallucinator (NOH), which pinpoints the objects nearby each proposal with a Gaussian distribution, together with NOH-NMS, which dynamically eases the suppression for the space that might contain other objects with a high likelihood. Compared to Greedy-NMS, our method, as the state-of-the-art, improves by $3.9\%$ AP, $5.1\%$ Recall, and $0.8\%$ $\text{MR}^{-2}$ on CrowdHuman to $89.0\%$ AP and $92.9\%$ Recall, and $43.9\%$ $\text{MR}^{-2}$ respectively.
