Full search page →

Boxes, overlap and suppression

AdvancedDLCV 03 · Object Detection: From R-CNN to DETRInteractive plot

Description

When do overlapping detections represent one object? Inspect how greedy suppression makes that decision.

Try this: Move B onto A, then raise the IoU threshold. Find a setting that keeps both. Can the rule tell duplicates from two nearby objects?

Predict first

At the default positions, boxes A (score 0.92) and B (score 0.84) overlap with IoU about 0.55 and the IoU threshold is 0.5. How many of the three boxes survive?

Explanation

Answer the prediction first, or open the explanation when you are ready.

Show the explanation

B overlaps the higher-scoring A by more than the threshold, so it is suppressed. C is far from both and stays. Move B away or raise the threshold to keep it.

Lab · Boxes, overlap and suppression

Predict first

At the default positions, boxes A (score 0.92) and B (score 0.84) overlap with IoU about 0.55 and the IoU threshold is 0.5. How many of the three boxes survive?

IoU(A, B)
0.551
IoU(B, C)
0
Kept boxes
2

IoU = intersection area / union area. A suppressed box never suppresses another box; only previously kept boxes participate. A: kept · B: suppressed · C: kept.

Try this: Move B onto A, then raise the IoU threshold. Find a setting that keeps both. Can the rule tell duplicates from two nearby objects?

Three fixed-score boxes of the same class. First keep scores ≥ the score threshold; then visit boxes by descending score and suppress any box whose intersection-over-union with a kept box is strictly greater than the IoU threshold. Scores are toy inputs, not model accuracy.

Method sources

Open in the Playground →