mjq11302010044/RRPN

some questions about rotation_demo.py !

FakerYFX opened this issue · 1 comments

@mjq11302010044
if the CLASSES = ('chinese', 'japenese',"English","germany"),I want to get the different class's result, how can I modify the code below:
//Visualize detections for each class
CONF_THRESH = conf
NMS_THRESH = 0.3
for cls_ind, cls in enumerate(CLASSES[1:]):
cls_ind += 1 # because we skipped background
cls_boxes = boxes[:, 5 * cls_ind:5 * (cls_ind + 1)] # D
cls_scores = scores[:, cls_ind]
dets = np.hstack((cls_boxes,
cls_scores[:, np.newaxis])).astype(np.float32)
keep = rotate_gpu_nms(dets, NMS_THRESH) # D
dets = dets[keep, :]
//dets = dets[0:20]
//dets[:, 4] = dets[:, 4] * 0.45
dets[:, 2] = dets[:, 2] / cfg.TEST.GT_MARGIN
dets[:, 3] = dets[:, 3] / cfg.TEST.GT_MARGIN
results = write_result_ICDAR(
im_file,
dets,
CONF_THRESH,
result_dir,
im_height,
im_width)
return results

Better to use a for loop to get predictions of different classes according to the bbox score