python - Convert overlapping bounding boxes to heatmap using predicted probabilities

Question

I am doing object detection based on a sliding window algorithm. Based on the following code I am creating heat maps :

def add_heat(heatmap, bbox_list):
    # Iterate through list of bboxes
    for box in bbox_list:
        # Add += 1 for all pixels inside each bbox
        # Assuming each "box" takes the form ((x1, y1), (x2, y2))
        heatmap[box[0][1]:box[1][1], box[0][0]:box[1][0]] += 1

The code simply adds +1 where bounding boxes are overlapping. However, for each bounding box I saved the score from the model prediction (model.predict_proba from tensorflow keras). Can anyone help me to incorporate the score from the model prediction, instead of just adding +1 to create a heatmap? Thanks a lot!

Answer 1

等待大神答复