文件列表:

.travis.yml (657, 2021-12-28)
LICENSE (1090, 2021-12-28)
NOTICE (7862, 2021-12-28)
data (0, 2021-12-28)
exp (0, 2021-12-28)
experiments (0, 2021-12-28)
experiments\ctdet_coco_dla_1x.sh (430, 2021-12-28)
experiments\ctdet_coco_dla_2x.sh (717, 2021-12-28)
experiments\ctdet_coco_hg.sh (504, 2021-12-28)
experiments\ctdet_coco_resdcn101.sh (462, 2021-12-28)
experiments\ctdet_coco_resdcn18.sh (499, 2021-12-28)
experiments\ctdet_pascal_dla_384.sh (288, 2021-12-28)
experiments\ctdet_pascal_dla_512.sh (347, 2021-12-28)
experiments\ctdet_pascal_resdcn101_384.sh (371, 2021-12-28)
experiments\ctdet_pascal_resdcn101_512.sh (423, 2021-12-28)
experiments\ctdet_pascal_resdcn18_384.sh (354, 2021-12-28)
experiments\ctdet_pascal_resdcn18_512.sh (402, 2021-12-28)
experiments\ddd_3dop.sh (249, 2021-12-28)
experiments\ddd_sub.sh (251, 2021-12-28)
experiments\exdet_coco_dla.sh (418, 2021-12-28)
experiments\exdet_coco_hg.sh (460, 2021-12-28)
experiments\multi_pose_dla_1x.sh (401, 2021-12-28)
experiments\multi_pose_dla_3x.sh (699, 2021-12-28)
experiments\multi_pose_hg_1x.sh (452, 2021-12-28)
experiments\multi_pose_hg_3x.sh (743, 2021-12-28)
images (0, 2021-12-28)
images\16004479832_a748d55f21_k.jpg (136674, 2021-12-28)
images\17790319373_bd19b24cfc_k.jpg (149246, 2021-12-28)
images\18124840932_e42b3e377c_k.jpg (162421, 2021-12-28)
images\19064748793_bb942deea1_k.jpg (140388, 2021-12-28)
images\24274813513_0cfd2ce6d0_k.jpg (113241, 2021-12-28)
images\33823288584_1d21cf0a26_k.jpg (249406, 2021-12-28)
images\33887522274_eebd074106_k.jpg (122273, 2021-12-28)
images\34501842524_3c858b3080_k.jpg (234375, 2021-12-28)
images\NOTICE (917, 2021-12-28)
... ...

# kitti_eval `evaluate_object_3d_offline.cpp`evaluates your KITTI detection locally on your own computer using your validation data selected from KITTI training dataset, with the following metrics: - overlap on image (AP) - oriented overlap on image (AOS) - overlap on ground-plane (AP) - overlap in 3D (AP) Compile `evaluate_object_3d_offline.cpp` with dependency of Boost and Linux `dirent.h` (You should already have it under most Linux). Run the evalutaion by: ./evaluate_object_3d_offline groundtruth_dir result_dir Note that you don't have to detect over all KITTI training data. The evaluator only evaluates samples whose result files exist. ### Updates - June, 2017: * Fixed the bug of detection box filtering based on min height according to KITTI's note on 25.04.2017.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：