added figure to README

Author: Davidbon1

README.md

@@ -1,6 +1,20 @@
 # Masked Event Modelling: Self-Supervised Pretraining for Event Cameras
+Link to paper: https://arxiv.org/abs/2212.10368v1
 
-Event cameras asynchronously capture brightness changes with low latency, high temporal resolution, and high dynamic range. Deploying deep learning methods for classification or other tasks to these sensors typically requires large labeled datasets. However, annotation of event data is a costly and laborious process. To reduce the dependency on labeled event data, we introduce Masked Event Modeling (MEM), a self-supervised pretraining framework for events. Our method pretrains a neural network on unlabeled events, which can originate from any event camera recording. Subsequently, the pretrained model is finetuned on a downstream task, leading to a consistently improved performance on the task while requiring fewer labels. Our method outperforms the state-of-the-art object classification across three datasets, N-ImageNet, N-Cars, and N-Caltech101, increasing the top-1 accuracy of previous work by significant margins. We further find that MEM is even superior to supervised RGB-based pretraining when tested on real-world event data. Models pretrained with MEM exhibit improved label efficiency, especially in low data regimes, and generalize well to the dense task of semantic image segmentation.
+Event cameras asynchronously capture brightness changes with low latency, high temporal resolution, and high dynamic range. However, annotation of event data is a costly and laborious process, which limits the use of deep learning methods for classification and other semantic tasks with the event modality. To reduce the dependency on labeled event data, we introduce Masked Event Modeling (MEM), a self-supervised framework for events. Our method pretrains a neural network on unlabeled events, which can originate from any event camera recording. Subsequently, the pretrained model is finetuned on a downstream task, leading to a consistent improvement of the task accuracy.
+For example, our method reaches state-of-the-art classification accuracy across three datasets, N-ImageNet, N-Cars, and N-Caltech101, increasing the top-1 accuracy of previous work by significant margins. When tested on real-world event data, MEM is even superior to supervised RGB-based pretraining. The models pretrained with MEM are also label-efficient and generalize well to the dense task of semantic image segmentation.
+
+![MEM](mem-pipeline-figure.png)
+
+Citation:
+```
+@inproceedings{masked-event-modeling,
+      title={Masked Event Modeling: Self-Supervised Pretraining for Event Cameras}, 
+      author={Simon Klenk, David Bonello, Lukas Koestler, Nikita Araslanov and Daniel Cremers},
+      booktitle=WACV,
+      year={2024}
+}
+```
 
 ## Setup
 
@@ -97,7 +111,7 @@ cd mem
 Follow the README in `mem/semantic_segmentation` to train a semantic segmentation model.
 
 
-## Atribution
+## Attribution
 
 - This code is based on Microsoft's BEiT code (https://github.com/microsoft/unilm/tree/master/beit).
 - The eventVAE code is based on Phil Wang's code (https://github.com/lucidrains/DALLE-pytorch)