deep learning
关注
分享
扫一扫
fourye007
Work for fun, Live for love
展开
-
Monocular 3D Object Detection and Box Fitting Trained End-to-End Using Intersection-over-Union Loss
Monocular 3D Object Detection and Box Fitting Trained End-to-End Using Intersection-over-Union Lossinsights3DIou lossHomoscedastic noiseHeteroscedastic noise3D框的representation 重要结论...原创 2019-11-14 17:24:03 · 638 阅读 · 0 评论 -
M3D-RPN
M3D-RPNPost 3D–>2D Optimization下图是M3D-RPN优化旋转角的方法其优化角度的方式挺不错的,严格利用了3d框投影在图像上的8个角点会与2D框重合该优化旋转角的方式,是否适合depth估计的迭代优化,即大致流程如下:d_init = predict()d_init_1 = d_init + sigmad_init_2 = d_init - sigm...原创 2019-10-30 21:08:50 · 1461 阅读 · 0 评论