Anchor-free Detection专题 Lejian Ren 20190512

为什么要有anchor 直接预测坐标难以训练 F(x) =(x1, y1, x2, y2)

早期的anchor-free: DenseBox FCN形式，输出5通道（score + 4个坐标变换） 对物体size差距较大的图像效果不好（FPN?） Huang, Lichao, et al. "Densebox: Unifying landmark localization with end to end object detection." arXiv preprint arXiv:1509.04874 (2015).

Denet 预测每个像素是什么类型的corner DeNet Tychsen-Smith, Lachlan, and Lars Petersson. "Denet: Scalable real-time object detection with directed sparse sampling." Proceedings of the IEEE International Conference on Computer Vision. 2017.

CornerNet Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet 为什么不用anchor 为什么用corner anchor多 超参多 Center比较难定位，需要考虑和四个边的关系 Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet Heatmap参考了key point Embedded是一个d-1的数 Offset用来减小从feature map到原图的影响（共享） Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet 预测corner 高斯 Focal loss Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet Corner pooling 为什么需要：物体的corner往往不在物体里面 Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet Grouping Corner Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet Law, Hei, and Jia Deng. "Cornernet: Detecting objects as paired keypoints." Proceedings of the European Conference on Computer Vision (ECCV). 2018.

CornerNet-Lite 加速 CornerNet-Saccade CornerNet-Squeeze 减小输入尺寸 利用attention减少需要预测的点 CornerNet-Squeeze 修改backbone Law, Hei, et al. "CornerNet-Lite: Efficient Keypoint Based Object Detection." arXiv preprint arXiv:1904.08900 (2019).

CornerNet-Lite CornerNet-Saccade Law, Hei, et al. "CornerNet-Lite: Efficient Keypoint Based Object Detection." arXiv preprint arXiv:1904.08900 (2019).

CornerNet-Lite Law, Hei, et al. "CornerNet-Lite: Efficient Keypoint Based Object Detection." arXiv preprint arXiv:1904.08900 (2019).

CenterNet-1 预测key point 预测长宽 X. Zhou, D. Wang, and P. Krähenbühl, “Objects as Points,” 2019.

CenterNet-1 Loss functions X. Zhou, D. Wang, and P. Krähenbühl, “Objects as Points,” 2019.

CenterNet-1 X. Zhou, D. Wang, and P. Krähenbühl, “Objects as Points,” 2019.

CenterNet-2 CenterNet的缺点 从预测corner变成预测corner+center 原文认为：只预测corner无法很好的catch物体本身的信息 我认为：embedding没有区分度 从预测corner变成预测corner+center 只有预测的center在预测的corner内部，才是正确的预测 Duan, Kaiwen, et al. "CenterNet: Keypoint Triplets for Object Detection." arXiv preprint arXiv:1904.08189 (2019).

CenterNet-2 Duan, Kaiwen, et al. "CenterNet: Keypoint Triplets for Object Detection." arXiv preprint arXiv:1904.08189 (2019).

CenterNet-2 如何判断center在corner内部（判定区域的设定） 太大，precision低 太小，recall低 自适应 Duan, Kaiwen, et al. "CenterNet: Keypoint Triplets for Object Detection." arXiv preprint arXiv:1904.08189 (2019).

CenterNet-2 Center pooling Cascade corner pooling Duan, Kaiwen, et al. "CenterNet: Keypoint Triplets for Object Detection." arXiv preprint arXiv:1904.08189 (2019).

CenterNet-2 Duan, Kaiwen, et al. "CenterNet: Keypoint Triplets for Object Detection." arXiv preprint arXiv:1904.08189 (2019).

RepPoints（representative points） motivation： 物体空间形状的可以通过几个reppoints来表示，通过学多个reppoints， 就可以用其外接矩形来表示物体的框 Z. Yang, S. Liu, H. Hu, L. Wang, and S. Lin, “RepPoints: Point Set Representation for Object Detection,” 2019.

RepPoints 和deformable convolution 完美结合 Z. Yang, S. Liu, H. Hu, L. Wang, and S. Lin, “RepPoints: Point Set Representation for Object Detection,” 2019.

RepPoints Pipeline： Learning RepPoints RepPoints refinement Converting RepPoints to bounding box Min-max function Partial min-max function Moment-based function (reppoints均值和二阶矩) Z. Yang, S. Liu, H. Hu, L. Wang, and S. Lin, “RepPoints: Point Set Representation for Object Detection,” 2019.

RepPoints Z. Yang, S. Liu, H. Hu, L. Wang, and S. Lin, “RepPoints: Point Set Representation for Object Detection,” 2019.

RepPoints Z. Yang, S. Liu, H. Hu, L. Wang, and S. Lin, “RepPoints: Point Set Representation for Object Detection,” 2019.

一点看法 为什么object detection没有一开始就anchor-free？ 目前的anchor-free就是“anchor-free”吗？ Anchor-free可能的发展方向