simonmeister/UnFlow
UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Loss
PythonMIT
Issues
- 10
tensorflow.python.framework.errors_impl.NotFoundError: ./backward_warp_op.so: undefined symbol: __cudaPushCallConfiguration
#56 opened by MrYANG23 - 0
Preprocessing KITTI raw data
#75 opened by xc-bnu - 2
error : .\backward_warp_op.so not found
#67 opened by yoryory - 1
Code related to evaluation metric
#74 opened by Pixie8888 - 0
Some confusion about census loss
#73 opened by hello7623 - 1
- 6
Do you write UnFlow using pytorch?
#42 opened by hzk7287 - 2
Any template to train new dataset?
#68 opened by Kindpire - 3
lib_handle = py_tf.TF_LoadLibrary(library_filename) tensorflow.python.framework.errors_impl.NotFoundError: HOW TO GET RIDE OF THIS ERROR
#69 opened by Fazankabir - 1
KeyError: "correlation"
#71 opened by David3310273 - 1
unable to train
#72 opened by harikiran2995 - 10
- 7
How flow vectors are stored and why do we need to do addition in this line of forward warping?
#65 opened by alisaaalehi - 0
Some advices about Sintel loss
#55 opened by JGyoung33 - 2
fine tuning pwc trained model
#66 opened by yoryory - 1
How to train with my downloaded data ?
#70 opened by RuiFeiHe - 1
- 4
Where does the list in folder 'kitti_excludes' come from ? Especially the list of kitti2012, which can't be found on official website.
#49 opened by sunshinezhe - 5
Doesn't work if cuda is not installed in usr/local
#39 opened by djl11 - 4
NoneType in downsample
#59 opened by Freephi - 0
- 0
Output and input node name of UnFlow.
#63 opened by shihao1104 - 2
Unsupervised training questions
#62 opened - 0
"step" parameter to load frames has no effect
#60 opened by eldar - 0
- 0
- 3
- 2
- 0
Intended behaviour of np.roll?
#57 opened by johannes-graeter - 3
Hardware requirements
#54 opened by johannes-graeter - 2
Pyramidal Decomposition and Downsampling
#36 opened by gsaibro - 2
Evaluation results compared with the paper
#52 opened by jytime - 2
converge too slow
#47 opened by SmartPorridge - 4
- 1
- 1
- 1
- 5
Check failed: dim.IsSet() Internal error: Got nullptr for Dimension in downsample
#34 opened by lenazherdeva - 11
- 3
- 1
Out of date download URLs
#38 opened by djl11 - 3
ResourceExhaustedError - conv6
#32 opened by lenazherdeva - 1
- 3
- 5
When finetuning, previous networks are kept fixed?
#31 opened by Blcony - 2
Why require an even number of frames?
#33 opened by clauslang - 3
- 2
- 2
- 1
If no ground truth is available, can self.eval(1) in run() in train.py be safely outcommented?
#26 opened by Santana18