Faster R-CNN: How to avoid multiple detections in same area?

Question

I use the Tensorflow object detection API to train the Pascal VOC dataset from scratch. I just had a look on the first results after 200k training steps and the results are okay, despite that I often have many detections of the same class in Overlapping regions. For example consider the following detections (ignore the wrong person detection in the first image):

Is there a general way to avoid such multiple detecions of the same object? I guess this is caused by overlapping Region proposals for which the Detection network predicts objects that fit the groundtruth data above the 0.7 IoU threshold, so maybe it would help to set this threshold a bit higher?

Answer 1

Do you mean literally from scratch, or did you use a fine_tune_checkpoint?

As others have suggested, you can increase the IOU threshold for NMS (decreasing it to something like .01 will effectively prevent a region from being detected if it overlaps with another).