diff --git a/README.md b/README.md
index 70cdf77..2c2f516 100644
--- a/README.md
+++ b/README.md
@@ -61,11 +61,11 @@ print(f"Coverage Ratio: {coverage_ratio:.2f}, Has NaN Values: {has_nan_values}")
 
 The grey cells represent the NaN locations, the blue ones represent the valid data, and the red ones represent the rows and columns removed by the algorithm:
 
-<img src="docs/source/_static/example0.png" width="400">
+<img src="https://github.com/CyrilJl/OptiMask/blob/main/docs/source/_static/example0.png?raw=true" width="400">
 
 OptiMask’s algorithm is useful for handling unstructured NaN patterns, as shown in the following example:
 
-<img src="docs/source/_static/example2.png" width="400">
+<img src="https://github.com/CyrilJl/OptiMask/blob/main/docs/source/_static/example2.png?raw=true" width="400">
 
 ## Documentation