Differences

This shows you the differences between two versions of the page.

--- documentation:vision:pose_estimation [2022/03/22 10:08] – robberth
+++ documentation:vision:pose_estimation [2022/09/02 14:04] (current) – external edit 127.0.0.1
@@ Line 4: / Line 4: @@
   * https://github.com/DLR-RM/AugmentedAutoencoder
   * https://github.com/shbe-aau/multi-pose-estimation
+  * https://github.com/HampusAstrom/yolov3 (reduced, simpler repo to get things working on heron)
+  * https://github.com/robberthofmanfm/yolo/releases (used as a fileserver for the weights to avoid git LFS)
-The former two are inspirations to the latter, which is the one we use. The authors are  [[hampus.astrom@cs.lth.se|Hampus Astrom]] and [[shbe@create.aau.dk|Stefan Bengtson]].
+The former two are inspirations to the third one. The authors are  [[hampus.astrom@cs.lth.se|Hampus Astrom]] and [[shbe@create.aau.dk|Stefan Bengtson]].
 It uses an autoencoder from pytorch3d, which itself is built on top of pytorch.
@@ Line 12: / Line 14: @@
 ==== Troubleshooting ====
-Since it runs in a docker container and uses nvidia, you'll want to pass the ''--gpus all'' flag to ''docker run''. However, when calling ''docker build'', no such flag exists. The build will fail.
+Since it runs in a docker container and uses nvidia, you'll want to pass the ''--gpus all'' flag to ''docker run''. \\
-For me, the answer was to specify the default docker runtime as one that uses nvidia-docker.
+However, when calling ''docker build'', no such flag exists. The build will fail. \\
-  - ''sudo apt-get install nvidia-container-runtime''
+For me, the answer was to specify the default docker runtime as one that uses nvidia-docker.\\
-  - edit ''/etc/docker/daemon.json'' and make sure it look something like this: <code> {
+  - ''sudo apt-get install nvidia-container-runtime''  # follow this guide https://nvidia.github.io/nvidia-container-runtime/
+  - edit ''/etc/docker/daemon.json'' and make sure it looks something like this: <code> {
     "runtimes": {
         "nvidia": {