chuchienshu / video-snapshot Goto Github PK
View Code? Open in Web Editor NEWThe official implementation of Video Snapshot: Single Image Motion Expansion via Invertible Motion Embedding. In TPAMI, 2020.
License: MIT License
The official implementation of Video Snapshot: Single Image Motion Expansion via Invertible Motion Embedding. In TPAMI, 2020.
License: MIT License
In the paper, you mentioned an interpolation module after the decoder. I was looking through the model.py
code and cannot seem to find the implementation for this. The def decode(latents, out_channels, is_train=False, reuse=False):
function seems to only be the decoder itself (before the interpolation).
Is my understanding correct? Is there an implementation of this module somewhere in the repo, or could you share the implementation for this? Thank you!
In the paper, you stated that the embedding loss includes a threshold matrix with \theta = 90.
May I know if this 90 means pixel value in the range [0,255] or a
constant element-wise value 90 despite any tensor normalization?
Hi all,
I got this error when running under the decode mode:
Caused by op 'decode/out/Conv2D', defined at:
File "inference.py", line 116, in <module> evaluate( checkpoint_dir, RUN_Encoder=False) File "inference.py", line 53, in evaluate
restored_imgs = decode(embed_batch, out_channels, is_train=False, reuse=False)
File "/disk1/user1/research/snapshot/Video-Snapshot/model.py", line 187, in decode
n = Conv2d(n, out_channels, filter_size=1, strides=1, padding='SAME', name='out')
File "/disk1/user1/research/snapshot/Video-Snapshot/model.py", line 13, in Conv2d
conv = tf.nn.conv2d(batch_input, filters, [1, strides, strides, 1], padding=padding)
File "/home/user1/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/ops/gen_nn_ops.py", line 631, in conv2d data_format=data_format, dilations=dilations, name=name)
File "/home/user1/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/home/user1/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 3271, in create_op op_def=op_def) File "/home/user1/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1650, in __init__
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access
InternalError (see above for traceback): Blas SGEMM launch failed : m=408960, n=27, k=256
[[Node: decode/out/Conv2D = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 1, 1, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:GPU:0"](decode/n256s1/2/Conv2D, decode/out/filter/read)]]
But running under embedding mode is okay. Could you help with this issue? Thanks!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.