hide Tracker mode until multi-threaded WASM lands

README.md

@@ -13,6 +13,6 @@ short_description: Open-vocabulary detection in-browser via WebGPU
 
 Interactive demo of [tiiuae/Falcon-Perception](https://huggingface.co/tiiuae/Falcon-Perception), a 0.6B-parameter vision-language model for open-vocabulary detection.
 
-Three render modes (Detection / Segment / Tracker). Video + webcam modes route detections through an IoU tracker with motion extrapolation so overlays animate at 60 fps between inference cycles.
+Image, webcam, and video modes with IoU-based multi-object tracking. Two render modes (Detection / Segment). A HUD "Tracker" mode is also built-in but disabled pending multi-threaded inference on HF Spaces.
 
 Weights: [shreyask/falcon-perception-onnx-webgpu](https://huggingface.co/shreyask/falcon-perception-onnx-webgpu).

index.html

@@ -15,7 +15,7 @@
         });
       }
     </script>
-    <script type="module" crossorigin src="/assets/index-D4Z-c1nG.js"></script>
+    <script type="module" crossorigin src="/assets/index-2MfOTjEy.js"></script>
     <link rel="stylesheet" crossorigin href="/assets/index-DpjQuLKn.css">
   </head>
   <body class="bg-gray-950 text-gray-100 antialiased">