Update README.md

PINTO0309 · Jan 19, 2025 · e31d513 · e31d513
1 parent 9af19a5
commit e31d513
Showing 1 changed file with 2 additions and 0 deletions.
diff --git a/462_Gaze-LLE/README.md b/462_Gaze-LLE/README.md
@@ -2,6 +2,8 @@
 
 Gaze-LLE provides a streamlined gaze architecture that learns only a lightweight gaze decoder on top of a frozen, pretrained visual encoder (DINOv2). Gaze-LLE learns 1-2 orders of magnitude fewer parameters than prior works and doesn't require any extra input modalities like depth and pose!
 
+---
+
 The reason the processing speed of the demo video appears to be around 30ms is because the heatmap rendering process using Pillow is slow. If you stop the heatmap rendering process by pressing the `A` key on the keyboard, the processing speed will increase to around 17ms. Of the 17ms, 10ms is the inference time for YOLOv9-E.
 
 - Single person test - `gazelle_dinov2_vitb14_inout_1x3x448x448_1xNx4.onnx` + ONNX-TensorRT