README modify

README.md

@@ -9,10 +9,6 @@ tags:
 
 Paper: [Segformer++: Efficient Token-Merging Strategies for High-Resolution Semantic Segmentation](https://arxiv.org/abs/2405.14467)
 
-![image](docs/figures/segmentation.png)
-
-![image](docs/figures/pose.png)
-
 ## Abstract
 
 Utilizing transformer architectures for semantic segmentation of high-resolution images is hindered by the attention's quadratic computational complexity in the number of tokens. A solution to this challenge involves decreasing the number of tokens through token merging, which has exhibited remarkable enhancements in inference speed, training efficiency, and memory utilization for image classification tasks. In this paper, we explore various token merging strategies within the framework of the SegFormer architecture and perform experiments on multiple semantic segmentation and human pose estimation datasets. Notably, without model re-training, we, for example, achieve an inference acceleration of 61% on the Cityscapes dataset while maintaining the mIoU performance. Consequently, this paper facilitates the deployment of transformer-based architectures on resource-constrained devices and in real-time applications.
@@ -79,8 +75,9 @@ The weights of the Segformer (Original) model were used to get the inference res
 
 **Step 1.** Clone Repository
 
-git clone git clone https://huggingface.co/TimM77/SegformerPlusPlus
-
+```shell
+git clone https://huggingface.co/TimM77/SegformerPlusPlus
+```
 
 **Step 2.** Install required Packets
 
@@ -92,10 +89,14 @@ pip install .
 **Step 3.** Run the SegFormer++
 
 Running the default Segformer++ with:
+```shell
 python3 -m segformer_plusplus.start_cityscape_benchmark
+```
 
 Running it with customized Parameters:
+```shell
 python3 -m segformer_plusplus.start_cityscape_benchmark --backbone [b1-b5] --head [bsm_hq, bsm_fast, n2d_2x2] --checkpoint [Path/To/Checkpoint]
+```
 
 
 ## Citation