Andrew Jong
Andrew Jong
Hi. I'm trying to run this on the deepfashion dataset, but for some reason no segmentations nor neural body fitting gets generated whatsoever. I'm quite confused as to what I'm...
Suggestion to mark whether the code base is official or a reproduction. Makes a big difference the code base was actually used to produce the paper's results, or if it's...
I feel the code in its current state is quite cumbersome. Training is also cumbersome. Visdom isn't as good as Tensorboard. Thinking of rehauling the codebase to be neat and...
Preprocessing requires setting up 2 separate repositories. It's crazy. We should improve this. Ideas 1. Add the two preprocessing repositories as submodules so they're in a central location 2. Create...
Bug where non-vanilla GAN losses don't run. Currently crashes on these.
SwapNet says they copy face data Add this to the inference script. Add option for "--face_dir" that points to the root dir for cloth and texture of the original person....
The previous issue for this #9 was closed, though I still don't see any link to download pretrained models. Any update on this, please?
Just checking if this works with PT 0.4.1 as well, because the README only specifies 0.4. I ask because I noticed the PyTorch 0.4.1 release tag includes "Breaking changes". Thanks!
Hi, I'm looking at both your work and [ATEN](https://github.com/HCPLab-SYSU/ATEN). The conclusions from both your papers are _very_ similar. [Your paper](https://arxiv.org/abs/1808.00157) > In this paper, we presented a novel detection-free Part...
Hi Mr. Rosenbaum, Really nice work! I have a question from the forward() function in [EXAMPLES.md](https://github.com/cle-ros/RoutingNetworks/blob/master/EXAMPLES.md). > y, meta, actions = self._initialization(y, tasks=tasks) > y, meta, task_actions = self._per_task_assignment(y, meta,...