[weighted_box_fussion] - an alternative for `box_non_max_suppression`

[hardikdava]

Search before asking

• I have searched the Supervision issues and found no similar feature requests.

Description

Current object detection models removes overlapping boxes by nms which can reduce accuracy of the final results. It can be avoided by Weighted Box Fusion which can accepts all the prediction whether from a single model or multiple models.

Reference: Weighted Box Fusion
Original Implementation: ensemble-boxes

Use case

import supervision as sv

wbf = sv.WeightedBoxFusion()

res_a = model_a(image)
det_a = sv.Detection(res_a)

res_b = model_b(image)
det_b = sv.Detection(res_b)

wbf_detections = wbf([det_a, det_b])

Additional

class WeightedBoxFusion:

	def __init__(self):
		pass
	
	def __call__(self, detections: Union[Detections, List[Detections]]) -> Detections:
		pass
		result_detections = Detections(...)
		return result_detections 

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[SkalskiP]

@hardikdava I have yet to hear of this post-processing approach. Looks interesting. IS it better than NMS?

[hardikdava]

@SkalskiP yes, it is better in terms of accuracy but not in terms of speed. You can learn about it more and direct comparision on this blogpost: https://learnopencv.com/weighted-boxes-fusion/

[AlainPilon]

@hardikdava Did you implement the WBF? Even if Supervision doesnt accept the feature, I could use it for my current project.

[SkalskiP]

@hardikdava, if you have that implemented somewhere already, I'd love to see it. I know that you are probably super busy, but some of the open-source contributors may be able to add it to our codebase.

[AlainPilon]

I found this: https://github.com/ZFTurbo/Weighted-Boxes-Fusion which does most of the work, we just need the detections coordinates to be as % instead of absolute pixel values. The packages also does NMS which would be a feature overlap with the existing Supervision code. I dont know how we should handle this:

• extract just want we need for WBF, or
• replace the existing NMS code with the one in the package (and get a few extra features for free).

My working version of Supervision has drifted quite a bit from the current release. Anything I do on my side would probably not be mergeable in main.

[hardikdava]

@AlainPilon Yes, I took reference from that project. @SkalskiP I can not contribute to open source projects due to my company policy. but I haven't started working on it. It was just an idea. But from my knowledge, it does not make effect on accuracy. The detection accuracy can boost by using SAHI.

[SkalskiP]

I can not contribute to open-source projects due to my company policy.

Sad to hear it :/ We miss you!