rnsandeep
commented
5 years ago The dimensions of your image are altered. change it to (1, height, width, 3) instead of (1, 3, height, width).
You can do that with im = np.reshape(im, (im.shape[2], im.shape[0], im.shape[1])) after reading the image. I guess you are reading the image with libraries other than opencv. better use opencv then no need to reshape.
Hi guys, I ve run the demo.py and it got error, saying
"Traceback (most recent call last): File "D:/tf-faster-rcnn-master/tf-faster-rcnn-master/tools/demo.py", line 153, in
demo(sess, net, im_name)
File "D:/tf-faster-rcnn-master/tf-faster-rcnn-master/tools/demo.py", line 85, in demo
scores, boxes = im_detect(sess, net, im)
File "D:\tf-faster-rcnn-master\tf-faster-rcnn-master\tools..\lib\model\test.py", line 93, in imdetect
, scores, bbox_pred, rois = net.test_image(sess, blobs['data'], blobs['im_info'])
File "D:\tf-faster-rcnn-master\tf-faster-rcnn-master\tools..\lib\nets\network.py", line 478, in test_image
feed_dict=feed_dict)
File "C:\Users\PC\Anaconda3\envs\tf-faster-rcnn-master\lib\site-packages\tensorflow\python\client\session.py", line 877, in run
run_metadata_ptr)
File "C:\Users\PC\Anaconda3\envs\tf-faster-rcnn-master\lib\site-packages\tensorflow\python\client\session.py", line 1076, in _run
str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (1, 3, 600, 800) for Tensor 'Placeholder:0', which has shape '(1, ?, ?, 3)')'"
I have no idea what I was massing.
My machine is Windows 10 (64 bit) : Python 3.6
If anyone faced this error before, please share me how to solve this. Thank very much.