aboutsummaryrefslogtreecommitdiff
path: root/readme.md
diff options
context:
space:
mode:
authorFrederic Becker2023-11-02 11:11:36 +0100
committerGitHub2023-11-02 11:11:36 +0100
commit83f34f0502177fe540907c592743b637325b9aba (patch)
treec2f5b6da642552db3f045f7dcfb01a8814969a19 /readme.md
parentf8302ee886ef9b631f11a52900dac964a61350e1 (diff)
Update readme.md
Diffstat (limited to 'readme.md')
-rw-r--r--readme.md9
1 files changed, 8 insertions, 1 deletions
diff --git a/readme.md b/readme.md
index eeadc4b..6935708 100644
--- a/readme.md
+++ b/readme.md
@@ -3,6 +3,13 @@
<b>TL;DR:</b> Introducing Loci-Looped, an extension to Loci, with a parameterized percept gate that learns to slot-wise fuse imaginations with sensations. As a result, Loci learns to track objects through occlusions and to imagine their trajectory. See our paper for more details: [arxiv](https://arxiv.org/abs/2310.10372)
+
+
+https://github.com/CognitiveModeling/Loci-Looped/assets/60604502/2b6d5fcf-8eeb-47db-ab26-1a9e4a3a48f0
+
+
+
+
---
## Requirements
A suitable [conda](https://conda.io/) environment named `loci-l` can be created
@@ -50,4 +57,4 @@ python -m scripts.exec.train -cfg [path-to-config-file] -load [path-to-model-fil
```
## Acknowledgements
-MOT metrics are computed with py-motmetrics. The visual evaluation metrics script is borrowed from the [Slotformer](https://github.com/pairlab/SlotFormer) repository. \ No newline at end of file
+MOT metrics are computed with py-motmetrics. The visual evaluation metrics script is borrowed from the [Slotformer](https://github.com/pairlab/SlotFormer) repository.