def setup(args):
"""
Create configs and perform basic setups.
"""
TEST_NAME = 'usopen_nadal_test2'
register_coco_instances(TEST_NAME, {},
f"./data/{TEST_NAME}/annotations.json",
f"./data/{TEST_NAME}/images")
cfg = get_cfg()
add_pointrend_config(cfg)
cfg.merge_from_file(
"./projects/PointRend/configs/InstanceSegmentation/pointrend_rcnn_R_50_FPN_3x_coco.yaml"
)
cfg.DATASETS.TEST = (TEST_NAME, )
cfg.MODEL.WEIGHTS = "./models/pointrend_rcnn_R_50_FPN_3x.pkl" # initialize from model zoo
cfg.OUTPUT_DIR = f'./output/{TEST_NAME}_pointrend'
# cfg.merge_from_file(args.config_file)
# cfg.merge_from_list(args.opts)
cfg.freeze()
default_setup(cfg, args)
return cfg
def setup_cfg(args):
cfg = get_cfg()
cfg.DATALOADER.NUM_WORKERS = 0
add_pointrend_config(cfg)
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
return cfg
def setup_cfg(args):
cfg = get_cfg()
# cuda context is initialized before creating dataloader, so we don't fork anymore
cfg.DATALOADER.NUM_WORKERS = 0
add_pointrend_config(cfg)
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
return cfg
def setup(args):
"""
Create configs and perform basic setups.
"""
cfg = get_cfg()
add_pointrend_config(cfg)
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
default_setup(cfg, args)
return cfg
def setup(args):
"""
Create configs and perform basic setups.
"""
cfg = get_cfg()
add_pointrend_config(cfg)
add_point_sup_config(cfg)
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
default_setup(cfg, args)
# Setup logger for "point_sup" module
setup_logger(output=cfg.OUTPUT_DIR, distributed_rank=comm.get_rank(), name="point_sup")
return cfg
def setup(args):
"""
Create configs and perform basic setups.
"""
cfg = get_cfg()
add_pointrend_config(cfg)
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
# if args.eval_only:
# cfg.MODEL.WEIGHTS = "/root/detectron2/projects/PointRend/log_50_50/model_0014999.pth"
cfg.freeze()
default_setup(cfg, args)
return cfg
def test_PointRend_mask_head_tracing(self):
cfg = model_zoo.get_config(
"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x.yaml")
point_rend.add_pointrend_config(cfg)
cfg.MODEL.ROI_HEADS.IN_FEATURES = ["p2", "p3"]
cfg.MODEL.ROI_MASK_HEAD.NAME = "PointRendMaskHead"
cfg.MODEL.ROI_MASK_HEAD.POOLER_TYPE = ""
cfg.MODEL.ROI_MASK_HEAD.POINT_HEAD_ON = True
chan = 256
head = point_rend.PointRendMaskHead(
cfg,
{
"p2": ShapeSpec(channels=chan, stride=4),
"p3": ShapeSpec(channels=chan, stride=8),
},
)
def gen_inputs(h, w, N):
p2 = torch.rand(1, chan, h, w)
p3 = torch.rand(1, chan, h // 2, w // 2)
boxes = random_boxes(N, max_coord=h)
return p2, p3, boxes
class Wrap(nn.ModuleDict):
def forward(self, p2, p3, boxes):
features = {
"p2": p2,
"p3": p3,
}
inst = Instances((p2.shape[2] * 4, p2.shape[3] * 4))
inst.pred_boxes = Boxes(boxes)
inst.pred_classes = torch.zeros(inst.__len__(),
dtype=torch.long)
out = self.head(features, [inst])[0]
return out.pred_masks
model = Wrap({"head": head})
model.eval()
with torch.no_grad(), patch_builtin_len():
traced = torch.jit.trace(model, gen_inputs(302, 208, 20))
inputs = gen_inputs(100, 120, 30)
out_eager = model(*inputs)
out_trace = traced(*inputs)
self.assertTrue(torch.allclose(out_eager, out_trace))
from detectron2.data import MetadataCatalog
coco_metadata = MetadataCatalog.get("coco_2017_val")
# import PointRend project
from detectron2.projects import point_rend
# get the pretrained model
config_R50 = "projects/PointRend/configs/InstanceSegmentation/pointrend_rcnn_R_50_FPN_3x_coco.yaml"
config_R101 = "projects/PointRend/configs/InstanceSegmentation/pointrend_rcnn_R_101_FPN_3x_coco.yaml"
config_X101 = "projects/PointRend/configs/InstanceSegmentation/pointrend_rcnn_X_101_32x8d_FPN_3x_coco.yaml"
weights_R50 = "detectron2://PointRend/InstanceSegmentation/pointrend_rcnn_R_50_FPN_3x_coco/164955410/model_final_edd263.pkl"
weights_R101 = "detectron2://PointRend/InstanceSegmentation/pointrend_rcnn_R_101_FPN_3x_coco/28119983/model_final_3f4d2a.pkl"
weights_X101 = "detectron2://PointRend/InstanceSegmentation/pointrend_rcnn_X_101_32x8d_FPN_3x_coco/28119989/model_final_ba17b9.pkl"
cfg = get_cfg()
point_rend.add_pointrend_config(cfg)
cfg.merge_from_file(config_R101)
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5
cfg.MODEL.WEIGHTS = weights_R101
predictor = DefaultPredictor(cfg)
def get_void_from_mask(person_mask_closed, height,
void_cri_top, void_cri_bottom):
top_void = 0
bottom_void = 0
for h_count in range(height):
line_mask_int = person_mask_clos[h_count, :, 0].astype(int)
if h_count < height * 0.5:
