memory_optimize=False yields no results

[Marvin1412]

Using the following code I only get emtpy polygons if memory_optimize is set to False. Otherwise I yield good results. Any Idea why? I'd also be interested what your POV is on the different paramters and what is optimal if I really try to not overlook something? Greetings and Thank you!

element_crops = MakeCropsDetectThem(
imgsz =800,
image=left_image,
model_path="yolov8m-seg.pt",
segment=True,
show_crops=False,
shape_x=400,
shape_y=400,
overlap_x=10,
overlap_y=10,
conf=0.5,
iou=0.9,
memory_optimize=False,
classes_list = list(range(80)),
resize_initial_size=True,
)
result = CombineDetections(element_crops, nms_threshold=0.5, match_metric='IOS')

visualize_results(
img=result.image,
confidences=result.filtered_confidences,
boxes=result.filtered_boxes,
polygons=result.filtered_polygons,
classes_ids=result.filtered_classes_id,
classes_names=result.filtered_classes_names,
segment=True,
thickness=8,
font_scale=1.1,
fill_mask=True,
show_boxes=False,
delta_colors=3,
show_class=False,
axis_off=False
)

[Koldim2001]

Try:

visualize_results(
img=result.image,
confidences=result.filtered_confidences,
boxes=result.filtered_boxes,
masks=result.filtered_masks,
classes_ids=result.filtered_classes_id,
classes_names=result.filtered_classes_names,
segment=True,
thickness=8,
font_scale=1.1,
fill_mask=True,
show_boxes=False,
delta_colors=3,
show_class=False,
axis_off=False
)

The reason is that when working with a more accurate mask inference mode, all processing is carried out in masks, not polygons (therefore, you need to feed the visualization function masks=result.filtered_masks)

#9

[Marvin1412]

Thank you for the Answer!