Chengfang.LU/HSAP
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: initial HSAP platform

Browse Source 
Huaxu Sentinel Active Safety Platform with embedded algorithm code,
Docker Compose setup, and vendored dataset scaffolds for clone-and-run.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
Chengfang Lu
2026-05-25 16:59:59 +08:00
commit 7c43b44c57
1619 changed files with 373355 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

83
algorithms/lane_ufld/code/pytorch-auto-drive-master/tools/vis/lane_img_dir.py Normal file
View File

@@ -0,0 +1,83 @@
import argparse
import torch
from importmagician import import_from
with import_from('./'):
from utils.args import read_config, parse_arg_cfg, cmd_dict, add_shortcuts
from utils.runners import LaneDetDir
try:
from utils.common import warnings
except ImportError:
import warnings
if __name__ == '__main__':
# Settings
parser = argparse.ArgumentParser(description='PytorchAutoDrive lane directory vis', conflict_handler='resolve')
add_shortcuts(parser)
parser.add_argument('--config', type=str, help='Path to config file', required=True)
parser.add_argument('--image-path', type=str,
help='Image input path')
parser.add_argument('--save-path', type=str,
help='Result output path')
# Optional args/to overwrite configs
parser.add_argument('--mask-path', type=str,
help='Mask input path, if both mask & keypoint are None,'
'inference will be performed ')
parser.add_argument('--keypoint-path', type=str,
help='Keypoint input path (expect json/txt file in CULane format, [x, y]),'
'if both mask & keypoint are None, inference will be performed')
parser.add_argument('--gt-keypoint-path', type=str,
help='Ground truth keypoint input path (expect json/txt file in CULane format, [x, y]),'
'if both mask & keypoint are None, inference will be performed')
parser.add_argument('--image-suffix', type=str, default='.jpg',
help='Image file suffix')
parser.add_argument('--keypoint-suffix', type=str, default='.lines.txt',
help='Keypoint file suffix')
parser.add_argument('--gt-keypoint-suffix', type=str, default='.lines.txt',
help='Ground truth keypoint file suffix')
parser.add_argument('--mask-suffix', type=str, default='.png',
help='Segmentation mask file suffix')
parser.add_argument('--style', type=str, default='point',
help='Lane visualization style: point/line/bezier')
parser.add_argument('--metric', type=str, default='culane',
help='Lane eval metric when comparing with GT')
parser.add_argument('--pred', action='store_true',
help='Whether to predict from a model')
parser.add_argument('--mixed-precision', action='store_true',
help='Enable mixed precision training')
parser.add_argument('--use-color-pool', action='store_true',
help='Use a larger color pool for lane segmentation masks')
parser.add_argument('--cfg-options', type=cmd_dict,
help='Override config options with \"x1=y1 x2=y2 xn=yn\"')
group2 = parser.add_mutually_exclusive_group()
group2.add_argument('--continue-from', type=str,
help='[Deprecated] Continue training from a previous checkpoint')
group2.add_argument('--checkpoint', type=str,
help='Continue/Load from a previous checkpoint')
retain_args = ['mixed_precision', 'pred', 'metric',
'image_path', 'save_path', 'mask_path', 'keypoint_path', 'gt_keypoint_path',
'image_suffix', 'keypoint_suffix', 'gt_keypoint_suffix', 'mask_suffix', 'use_color_pool', 'style']
args = parser.parse_args()
# Parse configs and build model
if args.mixed_precision and torch.__version__ < '1.6.0':
warnings.warn('PyTorch version too low, mixed precision training is not available.')
if args.image_path is not None and args.save_path is not None:
assert args.image_path != args.save_path, "Try not to overwrite your dataset!"
cfg = read_config(args.config)
args, cfg = parse_arg_cfg(args, cfg)
cfg_runner_key = 'vis' if 'vis' in cfg.keys() else 'test'
for k in retain_args:
cfg[cfg_runner_key][k] = vars(args)[k]
if not cfg[cfg_runner_key]['pred']:
assert cfg[cfg_runner_key]['style'] != 'bezier', 'Must use --pred for style bezier!'
cfg['model'] = None
runner = LaneDetDir(cfg=cfg)
runner.run()
runner.clean()

55
algorithms/lane_ufld/code/pytorch-auto-drive-master/tools/vis/lane_video.py Normal file
View File

@@ -0,0 +1,55 @@
import argparse
import torch
from importmagician import import_from
with import_from('./'):
from utils.args import read_config, parse_arg_cfg, cmd_dict, add_shortcuts
from utils.runners import LaneDetVideo
try:
from utils.common import warnings
except ImportError:
import warnings
if __name__ == '__main__':
# Settings
parser = argparse.ArgumentParser(description='PytorchAutoDrive lane video vis', conflict_handler='resolve')
add_shortcuts(parser)
parser.add_argument('--config', type=str, help='Path to config file', required=True)
parser.add_argument('--video-path', type=str, required=True,
help='Video input path')
parser.add_argument('--save-path', type=str, required=True,
help='Result output path')
parser.add_argument('--style', type=str, default='point',
help='Lane visualization style: point/line/bezier')
# Optional args/to overwrite configs
parser.add_argument('--mixed-precision', action='store_true',
help='Enable mixed precision training')
parser.add_argument('--cfg-options', type=cmd_dict,
help='Override config options with \"x1=y1 x2=y2 xn=yn\"')
group2 = parser.add_mutually_exclusive_group()
group2.add_argument('--continue-from', type=str,
help='[Deprecated] Continue training from a previous checkpoint')
group2.add_argument('--checkpoint', type=str,
help='Continue/Load from a previous checkpoint')
retain_args = ['mixed_precision',
'video_path', 'save_path', 'style']
args = parser.parse_args()
# Parse configs and build model
if args.mixed_precision and torch.__version__ < '1.6.0':
warnings.warn('PyTorch version too low, mixed precision training is not available.')
cfg = read_config(args.config)
args, cfg = parse_arg_cfg(args, cfg)
cfg_runner_key = 'vis' if 'vis' in cfg.keys() else 'test'
for k in retain_args:
cfg[cfg_runner_key][k] = vars(args)[k]
cfg[cfg_runner_key]['pred'] = True # Must predict
runner = LaneDetVideo(cfg=cfg)
runner.run()
runner.clean()

68
algorithms/lane_ufld/code/pytorch-auto-drive-master/tools/vis/seg_img_dir.py Normal file
View File

@@ -0,0 +1,68 @@
import argparse
import torch
from importmagician import import_from
with import_from('./'):
from utils.args import read_config, parse_arg_cfg, cmd_dict, add_shortcuts
from utils.runners import SegDir
try:
from utils.common import warnings
except ImportError:
import warnings
if __name__ == '__main__':
# Settings
parser = argparse.ArgumentParser(description='PytorchAutoDrive semantic segmentation directory vis',
conflict_handler='resolve')
add_shortcuts(parser)
parser.add_argument('--config', type=str, help='Path to config file', required=True)
parser.add_argument('--image-path', type=str, required=True,
help='Image input path')
parser.add_argument('--save-path', type=str, required=True,
help='Result output path')
# Optional args/to overwrite configs
parser.add_argument('--target-path', type=str,
help='Seg label input path, if both mask & keypoint are None,'
'inference will be performed ')
parser.add_argument('--image-suffix', type=str, default='_leftImg8bit.png',
help='Image file suffix')
parser.add_argument('--target-suffix', type=str, default='_gtFine_labelIds.png',
help='Segmentation mask file suffix')
parser.add_argument('--pred', action='store_true',
help='Whether to predict from a model')
parser.add_argument('--map-id', action='store_true',
help='Enable label_id_map, usually for ground truth mask in Cityscapes')
parser.add_argument('--mixed-precision', action='store_true',
help='Enable mixed precision training')
parser.add_argument('--cfg-options', type=cmd_dict,
help='Override config options with \"x1=y1 x2=y2 xn=yn\"')
group2 = parser.add_mutually_exclusive_group()
group2.add_argument('--continue-from', type=str,
help='[Deprecated] Continue training from a previous checkpoint')
group2.add_argument('--checkpoint', type=str,
help='Continue/Load from a previous checkpoint')
retain_args = ['mixed_precision', 'pred',
'image_path', 'save_path', 'target_path',
'image_suffix', 'target_suffix', 'map_id']
args = parser.parse_args()
# Parse configs and build model
if args.mixed_precision and torch.__version__ < '1.6.0':
warnings.warn('PyTorch version too low, mixed precision training is not available.')
assert args.image_path != args.save_path, "Try not to overwrite your dataset!"
cfg = read_config(args.config)
args, cfg = parse_arg_cfg(args, cfg)
cfg_runner_key = 'vis' if 'vis' in cfg.keys() else 'test'
for k in retain_args:
cfg[cfg_runner_key][k] = vars(args)[k]
if not cfg[cfg_runner_key]['pred']:
cfg['model'] = None
runner = SegDir(cfg=cfg)
runner.run()
runner.clean()

54
algorithms/lane_ufld/code/pytorch-auto-drive-master/tools/vis/seg_video.py Normal file
View File

@@ -0,0 +1,54 @@
import argparse
import torch
from importmagician import import_from
with import_from('./'):
from utils.args import read_config, parse_arg_cfg, cmd_dict, add_shortcuts
from utils.runners import SegVideo
try:
from utils.common import warnings
except ImportError:
import warnings
if __name__ == '__main__':
# Settings
parser = argparse.ArgumentParser(description='PytorchAutoDrive semantic segmentation video vis',
conflict_handler='resolve')
add_shortcuts(parser)
parser.add_argument('--config', type=str, help='Path to config file', required=True)
parser.add_argument('--video-path', type=str, required=True,
help='Video input path')
parser.add_argument('--save-path', type=str, required=True,
help='Result output path')
# Optional args/to overwrite configs
parser.add_argument('--mixed-precision', action='store_true',
help='Enable mixed precision training')
parser.add_argument('--cfg-options', type=cmd_dict,
help='Override config options with \"x1=y1 x2=y2 xn=yn\"')
group2 = parser.add_mutually_exclusive_group()
group2.add_argument('--continue-from', type=str,
help='[Deprecated] Continue training from a previous checkpoint')
group2.add_argument('--checkpoint', type=str,
help='Continue/Load from a previous checkpoint')
retain_args = ['mixed_precision',
'video_path', 'save_path']
args = parser.parse_args()
# Parse configs and build model
if args.mixed_precision and torch.__version__ < '1.6.0':
warnings.warn('PyTorch version too low, mixed precision training is not available.')
cfg = read_config(args.config)
args, cfg = parse_arg_cfg(args, cfg)
cfg_runner_key = 'vis' if 'vis' in cfg.keys() else 'test'
for k in retain_args:
cfg[cfg_runner_key][k] = vars(args)[k]
cfg[cfg_runner_key]['pred'] = True # Must predict
runner = SegVideo(cfg=cfg)
runner.run()
runner.clean()