Functions — github.com/NVIDIA-AI-IOT/CUDA-PointPillars

CodeHub MCPcopy Create free account

hub / github.com/NVIDIA-AI-IOT/CUDA-PointPillars / functions

Functions203 in github.com/NVIDIA-AI-IOT/CUDA-PointPillars

⨍ Functions 203◇ Types & classes 30

↓ 10 callersFunctionprint_str

(value, *arg, sstream=None)

tool/eval/eval.py:710

↓ 8 callersFunctionget_mAP_v2

tool/eval/eval.py:602

↓ 6 callersFunction_extend_matrix

tool/eval/kitti_common.py:55

↓ 6 callersMethodforward

src/pointpillar/pointpillar.cpp:127

↓ 6 callersMethodload

src/common/tensorrt.cpp:168

↓ 6 callersMethodsize

src/common/tensor.hpp:105

↓ 4 callersFunctioneval_class

Kitti eval. support 2d/bev/3d/aos eval. support 0.5:0.05:0.95 coco AP. Args: gt_annos: dict, must from get_label_annos() in kitti_common.p

tool/eval/eval.py:481

↓ 4 callersFunctionget_kitti_info_path

(idx, prefix, info_type='image_2', fil

tool/eval/kitti_common.py:14

↓ 4 callersFunctionlidar_to_top_coords

tool/eval/kitti_util.py:435

↓ 4 callersMethodstart

src/common/timer.hpp:43

↓ 4 callersMethodstop

src/common/timer.hpp:45

↓ 3 callersMethodcart2hom

Input: nx3 points in Cartesian Oupput: nx4 points in Homogeneous by pending 1

tool/eval/kitti_util.py:213

↓ 3 callersFunctionhypothesis_func

tool/eval/kitti_util.py:770

↓ 3 callersMethodproject_velo_to_image

Input: nx3 points in velodyne coord. Output: nx2 points in image2 coord.

tool/eval/kitti_util.py:264

↓ 3 callersMethodproject_velo_to_rect

(self, pts_3d_velo)

tool/eval/kitti_util.py:247

↓ 3 callersMethodread_calib_file

Read in a calibration file and parse into a dictionary. Ref: https://github.com/utiasSTARS/pykitti/blob/master/pykitti/utils.py

tool/eval/kitti_util.py:174

↓ 2 callersFunctionarea

Computes area of boxes. Args: boxes: Numpy array with shape [N, 4] holding N boxes Returns: a numpy array with shape [N*1] r

tool/eval/kitti_common.py:348

↓ 2 callersFunctioncompute_statistics_jit

(overlaps, gt_datas, dt_datas,

tool/eval/eval.py:184

↓ 2 callersFunctiondata_type_string

src/common/tensorrt.cpp:81

↓ 2 callersMethodend

src/common/tensor.hpp:100

↓ 2 callersFunctionformat_shape

src/common/tensorrt.cpp:51

↓ 2 callersFunctionget_image_index_str

tool/eval/kitti_common.py:10

↓ 2 callersFunctionget_label_anno

tool/eval/kitti_common.py:293

↓ 2 callersFunctionget_split_parts

(num, num_part)

tool/eval/eval.py:305

↓ 2 callersFunctionimage_box_overlap

(boxes, query_boxes, criterion=-1)

tool/eval/eval.py:99

↓ 2 callersMethodlog

src/common/tensorrt.cpp:44

↓ 2 callersFunctionloop_node

(graph, current_node, loop_time=0)

tool/modify_onnx.py:34

↓ 2 callersFunctionproject_to_image

Project 3d points to image plane. Usage: pts_2d = projectToImage(pts_3d, P) input: pts_3d: nx3 matrix P: 3x4 projection

tool/eval/kitti_util.py:596

↓ 2 callersFunctionroty

Rotation about the y-axis.

tool/eval/kitti_util.py:347

↓ 2 callersFunctionsimplify_preprocess

tool/modify_onnx.py:77

↓ 1 callersFunctionGetDeviceInfo

src/main.cpp:29

↓ 1 callersFunctionSaveBoxPred

src/main.cpp:109

↓ 1 callersFunction_prepare_data

(gt_annos, dt_annos, current_class, difficulty)

tool/eval/eval.py:450

↓ 1 callersFunction_read_imageset_file

tool/eval/kitti_format.py:89

↓ 1 callersFunction_read_imageset_file

tool/eval/evaluate.py:7

↓ 1 callersFunctionadd_difficulty_to_annos

tool/eval/kitti_common.py:248

↓ 1 callersMethodbegin

src/common/tensor.hpp:95

↓ 1 callersFunctionbev_box_overlap

(boxes, qboxes, criterion=-1)

tool/eval/eval.py:128

↓ 1 callersFunctioncalculate_iou_partly

fast iou algorithm. this function can be used independently to do result analysis. Args: gt_annos: dict, must from get_label_annos()

tool/eval/eval.py:364

↓ 1 callersFunctionclean_data

(gt_anno, dt_anno, current_class, difficulty)

tool/eval/eval.py:39

↓ 1 callersFunctioncompute_box_3d

Takes an object and a projection matrix (P) and projects the 3d bounding box into the image plane. Returns: corners_2d: (

tool/eval/kitti_format.py:56

↓ 1 callersFunctioncompute_box_3d

Takes an object and a projection matrix (P) and projects the 3d bounding box into the image plane. Returns: corners_2d: (

tool/eval/transDet.py:62

↓ 1 callersMethodconstruct

src/common/tensorrt.cpp:158

↓ 1 callersFunctionconv

tool/eval/transDet.py:11

↓ 1 callersFunctioncreateNVOutput

tool/eval/kitti_format.py:94

↓ 1 callersFunctioncreateNVOutput

tool/eval/transDet.py:95

↓ 1 callersFunctioncreatePCDetOutput

tool/eval/kitti_format.py:179

↓ 1 callersFunctioncreate_core

src/main.cpp:134

↓ 1 callersFunctiond3_box_overlap

kitti camera format z_axis=1.

tool/eval/eval.py:172

↓ 1 callersFunctiond3_box_overlap_kernel

z_axis: the z (height) axis. z_center: unified z (height) center of box.

tool/eval/eval.py:134

↓ 1 callersFunctiondo_coco_style_eval

(gt_annos, dt_annos, current_classes, ove

tool/eval/eval.py:681

↓ 1 callersFunctiondo_eval_v2

(gt_annos, dt_annos, current_classes, min_overlaps,

tool/eval/eval.py:609

↓ 1 callersFunctiondo_eval_v3

(gt_annos, dt_annos, current_classes, min_overlaps,

tool/eval/eval.py:655

↓ 1 callersFunctiondump_fit_cost

(w_fit, train_x, train_y)

tool/eval/kitti_util.py:787

↓ 1 callersFunctiondump_fit_func

tool/eval/kitti_util.py:780

↓ 1 callersMethodempty

src/common/tensor.hpp:107

↓ 1 callersFunctionerror_func

(w, train_x, train_y)

tool/eval/kitti_util.py:776

↓ 1 callersMethodestimate_diffculty

Function that estimate difficulty to detect the object as defined in kitti website

tool/eval/kitti_util.py:84

↓ 1 callersMethodforward

src/common/tensorrt.cpp:194

↓ 1 callersMethodforward_only

src/pointpillar/pointpillar.cpp:71

↓ 1 callersMethodforward_timer

src/pointpillar/pointpillar.cpp:91

↓ 1 callersFunctionfused_compute_statistics

(overlaps, pr, gt_nums,

tool/eval/eval.py:315

↓ 1 callersFunctiongetFolderFile

src/main.cpp:61

↓ 1 callersFunctionget_calib_path

(idx, prefix, training=True, relative_path=True)

tool/eval/kitti_common.py:50

↓ 1 callersFunctionget_coco_eval_result

(gt_annos, dt_annos, current_classes,

tool/eval/eval.py:798

↓ 1 callersFunctionget_depth_pt3d

tool/eval/kitti_util.py:332

↓ 1 callersFunctionget_image_path

(idx, prefix, training=True, relative_path=True)

tool/eval/kitti_common.py:35

↓ 1 callersFunctionget_label_path

(idx, prefix, training=True, relative_path=True)

tool/eval/kitti_common.py:40

↓ 1 callersFunctionget_official_eval_result

gt_annos and dt_annos must contains following keys: [bbox, location, dimensions, rotation_y, score]

tool/eval/eval.py:718

↓ 1 callersFunctionget_thresholds

(scores: np.ndarray, num_gt, num_sample_pts=41)

tool/eval/eval.py:18

↓ 1 callersFunctionget_velodyne_path

(idx, prefix, training=True, relative_path=True)

tool/eval/kitti_common.py:45

↓ 1 callersFunctionhasEnding

src/main.cpp:52

↓ 1 callersFunctionhelp

src/main.cpp:170

↓ 1 callersMethodinit

src/pointpillar/pointpillar.cpp:42

↓ 1 callersFunctionintersection

Compute pairwise intersection areas between boxes. Args: boxes1: a numpy array with shape [N, 4] holding N boxes boxes2: a numpy

tool/eval/kitti_common.py:364

↓ 1 callersFunctioninverse_rigid_trans

Inverse a rigid body transform matrix (3x4 as [R|t]) [R'|-R't; 0|1]

tool/eval/kitti_util.py:368

↓ 1 callersFunctioniou

Computes pairwise intersection-over-union between box collections. Args: boxes1: a numpy array with shape [N, 4] holding N boxes.

tool/eval/kitti_common.py:395

↓ 1 callersFunctionloadData

src/main.cpp:80

↓ 1 callersFunctionload_file

src/common/tensorrt.cpp:63

↓ 1 callersFunctionmain

tool/export_onnx.py:86

↓ 1 callersFunctionparse_config

tool/export_onnx.py:71

↓ 1 callersMethodprint

src/common/tensorrt.cpp:251

↓ 1 callersMethodproject_8p_to_4p

tool/eval/kitti_util.py:271

↓ 1 callersMethodproject_image_to_rect

Input: nx3 first two channels are uv, 3rd channel is depth in rect camera coord. Output: nx3 points in rect camera coo

tool/eval/kitti_util.py:292

↓ 1 callersMethodproject_image_to_velo

(self, uv_depth)

tool/eval/kitti_util.py:306

↓ 1 callersMethodproject_rect_to_image

Input: nx3 points in rect camera coord. Output: nx2 points in image2 coord.

tool/eval/kitti_util.py:254

↓ 1 callersMethodproject_rect_to_ref

Input and Output are nx3 points

tool/eval/kitti_util.py:232

↓ 1 callersMethodproject_rect_to_velo

Input: nx3 points in rect camera coord. Output: nx3 points in velodyne coord.

tool/eval/kitti_util.py:240

↓ 1 callersMethodproject_ref_to_rect

Input and Output are nx3 points

tool/eval/kitti_util.py:236

↓ 1 callersMethodproject_ref_to_velo

(self, pts_3d_ref)

tool/eval/kitti_util.py:228

↓ 1 callersFunctionproject_to_image

Project 3d points to image plane. Usage: pts_2d = projectToImage(pts_3d, P) input: pts_3d: nx3 matrix P: 3x4 projection

tool/eval/kitti_format.py:28

↓ 1 callersFunctionproject_to_image

Project 3d points to image plane. Usage: pts_2d = projectToImage(pts_3d, P) input: pts_3d: nx3 matrix P: 3x4 projection

tool/eval/transDet.py:34

↓ 1 callersMethodproject_velo_to_ref

(self, pts_3d_velo)

tool/eval/kitti_util.py:224

↓ 1 callersMethodread_calib_from_video

Read calibration for camera 2 from video calib files. there are calib_cam_to_cam and calib_velo_to_cam under the calib_root_dir

tool/eval/kitti_util.py:194

↓ 1 callersFunctionroty

Rotation about the y-axis.

tool/eval/kitti_format.py:50

↓ 1 callersFunctionroty

Rotation about the y-axis.

tool/eval/transDet.py:56

↓ 1 callersMethodsetPluginNamespace

src/pointpillar/pointpillar-scatter.cpp:239

↓ 1 callersMethodset_timer

src/pointpillar/pointpillar.cpp:135

↓ 1 callersFunctionsimplify_postprocess

tool/modify_onnx.py:40

↓ 1 callersFunctionstartswith

src/main.cpp:158

next →1–100 of 203, ranked by callers