Update README.md

removed the image, did not look good.

README.md

@@ -16,12 +16,12 @@ Trained on synthetic PBR data only. Supports two evaluation modes:
 1. **Keypoints-only** (GT bounding box) -- upper bound on pose accuracy
 2. **Detection + Pose** (full pipeline, no GT) -- RTMDet detection, DOPER-t keypoints, PnP solve
 
-## Demo: Full Pipeline (Detection + Keypoints)
+<!-- ## Demo: Full Pipeline (Detection + Keypoints)
 
 ![DOPER-t detection + keypoints on HOPE val](demo/doper_hope_det_keypoints.png)
 
 *28 per-object RTMDet detectors + 28 per-object DOPER-t keypoint models running on a single HOPE val image. No GT bounding boxes.*
-
+ -->
 ## Results Summary
 
 ### Keypoints-Only (GT BBox)