Fixes

app.py

@@ -88,7 +88,7 @@ description = """
 <br/>a) <a href="https://huggingface.co/spaces/CVPR/regionclip-demo">RegionCLIP</a>
 <br/>b) <a href="https://colab.research.google.com/drive/19LBqQg0cS36rTLL_TaXZ7Ka9KJGkxiSe?usp=sharing">Colab notebook for Object-Centric-OVD</a>
 </div>
-<br/><div style=\"font-size:16px; color: #4f4f4f; text-align: left\">Note: While most examples showcased illustrate model capabilities, some illustrate model's limitations - such as finding globe,bird cage,teapot etc. in the picture etc. Also, the model appears to have text detection and recognition capabilities, even if text recognition is only very limited</div>
+<br/><div style=\"font-size:16px; color: #4f4f4f; text-align: left\"><b>Note: Inference time depends on input image size. Typically images with dimensions less than 500px has response time under 5 secs on CPU.</b><br/><i> While most examples showcased illustrate model capabilities, some illustrate model's limitations - such as finding globe,bird cage,teapot etc. in the picture etc. Also, the model appears to have text detection and recognition capabilities, even if text recognition is only very limited</i></div>
 <div style=\"font-size:14px; color: #6f6f6f; text-align: left\"><i>Images below are from&nbsp;&nbsp;<a href="https://en.wikipedia.org/wiki/Hidden_object_game">Wikipedia</a>,&nbsp;<a href="http://images.cocodataset.org/val2017/000000133819.jpg">COCO</a> and <a href="http://host.robots.ox.ac.uk/pascal/VOC/voc2012/">PASCAL VOC 2012</a>&nbsp;datasets </i></div>
 """
 demo = gr.Interface(