Update images

README.md

@@ -15,22 +15,22 @@ Build intelligence into your apps using machine learning models from the researc
 
 ### Image Classification
 
-![Image Classification Example](https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/FastViT_2x.png?download=true)
-
 - [FastViT](https://huggingface.co/collections/coreml-projects/coreml-fastvit-666b0053e54816747071d755), a small and very fast model family for image classification. 
 
-### Image Segmentation
+<img src="https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/FastViT_2x.png?download=true" width="50%">
 
-![Image Segmentation Example](https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/DETRSegmentation_2x.png?download=true)
+### Image Segmentation
 
 - [DETR](https://huggingface.co/coreml-projects/coreml-detr-semantic-segmentation). DEtection TRansformer (DETR) allows semantic segmentation on iOS devices, where each pixel in an image is classified according to the most probable category it belongs to.
 
-### Depth Estimation
+<img src="https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/DETRSegmentation_2x.png?download=true" width="50%">
 
-![Depth Estimation Example](https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/DepthAnything_2x.png?download=true)
+### Depth Estimation
 
 - [Depth Anything](https://huggingface.co/coreml-projects/coreml-depth-anything-small) is a state of the art model for monocular depth estimation.
 
+<img src="https://huggingface.co/datasets/coreml-projects/documentation-assets/resolve/main/DepthAnything_2x.png?download=true" width="50%">
+
 ### Text-to-Image Generation
 - [Stable Diffusion](https://huggingface.co/collections/apple/core-ml-stable-diffusion-666b3b0f4b5f3d33c67c6bbe) – Core ML versions of Stable Diffusion models for image generation on Apple Silicon.