Use 0.05 confidence threshold in icon_detect_v3 example to match icon_detect default

README.md

@@ -35,7 +35,7 @@ from torchvision.ops import nms
 
 IMGSZ = 1280
 STRIDES = (8, 16, 32)
-CONF_THRESHOLD = 0.35
+CONF_THRESHOLD = 0.05
 IOU_THRESHOLD = 0.45
 
 model = torch.jit.load("icon_detect_v3/model.pt", map_location="cpu").eval()
@@ -89,7 +89,7 @@ boxes[:, 1::2] = boxes[:, 1::2].clamp(0, height)
 
 Notes:
 - The model is single class (interactable region), so `cls` has one channel per scale.
-- `CONF_THRESHOLD` / `IOU_THRESHOLD` are starting points; tune them for your screenshots the same way you would tune `box_threshold` / `iou_threshold` for `icon_detect`.
+- `CONF_THRESHOLD` defaults to `0.05` to match the `box_threshold` that OmniParser uses with `icon_detect`. The detector is deliberately low confidence on small UI elements, so raising this much above `0.1` starts dropping real elements (icons, toolbar buttons, footer links) rather than just filtering noise. Tune `CONF_THRESHOLD` / `IOU_THRESHOLD` for your screenshots the same way you would tune `box_threshold` / `iou_threshold` for `icon_detect`.
 - The letterbox padding above is top-left anchored, which keeps the coordinate mapping to a single `scale` division. If you center the padding instead, subtract the pad offsets before dividing.
 
 # Responsible AI Considerations