search

hughw19 / NOCS_CVPR2019

[CVPR2019 Oral] Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation on Python3, Tensorflow, and Keras
https://geometry.stanford.edu/projects/NOCS_CVPR2019/
Other
450 stars 72 forks source link

no instance is detected #79

Open loubron23 opened 10 months ago

loubron23 commented 10 months ago

When I run the detect code, the problem of no instance is detected will appear, and an error will be reported when a certain image is detected: Image id: 52 data/real/test/scene_1/0054 data/gts/real_test/results_real_test_scene_1_0054.pkl predict result: 0.96875 0.96875 0.96875 Traceback (most recent call last): File "detect_eval.py", line 249, in <module> detect_result = model.detect([image], verbose=0) File "/rootCS/model.py", line 4394, in detect image.shape, windows[i]) File "/rootCS/model.py", line 4335, in unmold_detections full_mask = utils.unmold_mask(masks[i], boxes[i], image_shape) File "/rootCS/utils.py", line 1222, in unmold_mask mask, (y2 - y1, x2 - x1), interp='bilinear').astype(np.float32) / 255.0 File "/root/anaconda3/envs/py356/lib/python3.5/site-packages/numpy/lib/utils.py", line 101, in newfunc return func(*args, **kwds) File "/root/anaconda3/envs/py356/lib/python3.5/site-packages/scipy/misc/pilutil.py", line 565, in imresize imnew = im.resize(size, resample=func[interp]) File "/root/anaconda3/envs/py356/lib/python3.5/site-packages/PIL/Image.py", line 1929, in resize return self._new(self.im.resize(size, resample, box)) ValueError: height and width must be > 0 I checked that the input of the image is correct. I want to know what the problem is. Is it because my environment is configured incorrectly?

microchila commented 8 months ago

ME TOO, WHY????!!!!

loubron23 commented 8 months ago

ME TOO, WHY????!!!!

I still can't solve the problem. I guess it may be the environment problem(cuda, gpu)? If you solve it, please tell me!