How to simply test the model_kitti_stereo on left image and right image ?
WenguoLi opened this issue · comments
Wenguo.Li commented
Hi,
Thanks for your work!
I downloaded the model_kitti_stereo from your page, and modify the script monodepth_simple.py in the hope of a simple test.
32c32,33
< parser.add_argument('--image_path', type=str, help='path to the image', required=True)
---
> parser.add_argument('--left_image_path', type=str, help='path to the image', required=True)
> parser.add_argument('--right_image_path', type=str, help='path to the image', required=True)
52c53,54
< left = tf.placeholder(tf.float32, [2, args.input_height, args.input_width, 3])
---
> left = tf.placeholder(tf.float32, [1, args.input_height, args.input_width, 3])
> right = tf.placeholder(tf.float32, [1, args.input_height, args.input_width, 3])
55,59c57,67
< input_image = scipy.misc.imread(args.image_path, mode="RGB")
< original_height, original_width, num_channels = input_image.shape
< input_image = scipy.misc.imresize(input_image, [args.input_height, args.input_width], interp='lanczos')
< input_image = input_image.astype(np.float32) / 255
< input_images = np.stack((input_image, np.fliplr(input_image)), 0)
---
> left_image = scipy.misc.imread(args.left_image_path, mode="RGB")
> original_height, original_width, num_channels = left_image.shape
> left_image = scipy.misc.imresize(left_image, [args.input_height, args.input_width], interp='lanczos')
> left_image = left_image.astype(np.float32) / 255
> left_images = np.stack((left_image, np.fliplr(left_image)), 0)
>
> right_image = scipy.misc.imread(args.right_image_path, mode="RGB")
> original_height, original_width, num_channels = right_image.shape
> right_image = scipy.misc.imresize(right_image, [args.input_height, args.input_width], interp='lanczos')
> right_image = right_image.astype(np.float32) / 255
> right_images = np.stack((right_image, np.fliplr(right_image)), 0)
78,79c86,87
< disp = sess.run(model.disp_left_est[0], feed_dict={left: input_images})
< disp_pp = post_process_disparity(disp.squeeze()).astype(np.float32)
---
> disp = sess.run(model.disp_left_est[0], feed_dict={left: left_images, right: right_images})
> disp_pp = post_process_disparity(disp[0].squeeze()).astype(np.float32)
96,97c104,105
< batch_size=2,
< num_threads=1,
---
> batch_size=1,
> num_threads=2,
99c107
< do_stereo=False,
---
> do_stereo=True,
then, run by the following command
`
python monodepth_simple_stereo.py --left_image_path images/org_left.jpg --right_image_path images/org_right.jpg --checkpoint_path models/model_kitti_stereo/model_kitti_stereo
`
but errors occurred,
Traceback (most recent call last):
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 455, in _apply_op_helper
as_ref=input_arg.is_ref)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1211, in internal_convert_n_to_tensor
ctx=ctx))
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1146, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/constant_op.py", line 229, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/constant_op.py", line 208, in constant
value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/tensor_util.py", line 430, in make_tensor_proto
raise ValueError("None values not supported.")
ValueError: None values not supported.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "monodepth_simple_stereo.py", line 118, in <module>
tf.app.run()
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 125, in run
_sys.exit(main(argv))
File "monodepth_simple_stereo.py", line 115, in main
test_simple(params)
File "monodepth_simple_stereo.py", line 55, in test_simple
model = MonodepthModel(params, "test", left, None)
File "/home/apuser/deeplearning/tensorflow/examples/DepthEstimation/monodepth/monodepth_model.py", line 50, in __init__
self.build_model()
File "/home/apuser/deeplearning/tensorflow/examples/DepthEstimation/monodepth/monodepth_model.py", line 297, in build_model
self.model_input = tf.concat([self.left, self.right], 3)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/ops/array_ops.py", line 1124, in concat
return gen_array_ops.concat_v2(values=values, axis=axis, name=name)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/ops/gen_array_ops.py", line 1033, in concat_v2
"ConcatV2", values=values, axis=axis, name=name)
File "/home/apuser/tf_venv/tf_py3/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 483, in _apply_op_helper
raise TypeError("%s that don't all match." % prefix)
TypeError: Tensors in list passed to 'values' of 'ConcatV2' Op have types [float32, <NOT CONVERTIBLE TO TENSOR>] that don't all match.
SONG Ke commented
I am having the same issue. Could you help?