Maximum Matching Accuracy: An Instance Segmentation Evaluation Metric Utilizing Globally Optimal Matching
Quick Answer
The proposed Maximum Matching Accuracy (MMA) metric offers a threshold-free, continuous evaluation for instance segmentation, outperforming traditional metrics like AP@50 and PQ in stability and sensitivity.
Quick Take
The proposed Maximum Matching Accuracy (MMA) metric offers a threshold-free, continuous evaluation for instance segmentation, outperforming traditional metrics like AP@50 and PQ in stability and sensitivity. It addresses common issues in biological imaging, providing a reliable foundation for benchmarking segmentation models.
Key Points
- MMA finds globally optimal one-to-one matches between predicted and ground truth objects.
- It aggregates total overlap using per-pixel normalization for improved accuracy.
- MMA shows better stability and sensitivity compared to AP@50, PQ, and SEG metrics.
- The metric addresses issues like split and merged cells in biological imaging.
- MMA provides a principled approach for fair instance segmentation benchmarking.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 10107v1 Announce Type: new Abstract: Reliable evaluation of instance segmentation models requires metrics that accurately and consistently reflect segmentation quality.
However, the metrics most widely used in biological imaging carry fundamental mathematical weaknesses: hard Intersection-over-Union (IoU) thresholds that produce discontinuous, low sensitivity scoring; per-object normalization that distorts scores under object size variation; and greedy or one-to-many matching procedures that yield non-optimal, order-dependent correspondences.
Together, these properties produce unintuitive and unreliable model rankings under common failure modes such as split cells, merged cells, and cell boundary imprecision. We propose Maximum Matching Accuracy (MMA), a threshold-free continuous metric that finds a globally optimal one-to-one matching between predicted and ground truth objects and aggregates total overlap using per-pixel normalization.
We evaluate MMA against AP@50, PQ, SEG, and AJI across three experiments: synthetic failure cases, progressive corruption tests, and a model ranking comparison. MMA produces scores that are more stable, more sensitive, and more interpretable than existing alternatives, providing a principled foundation for fair instance segmentation benchmarking in biological cell imaging.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.