</> Code with Papers by Syntology

Papers › No time to train! Training-Free Reference-Based Instance Segmentation

No time to train! Training-Free Reference-Based Instance Segmentation

3 Jul 2025arXiv:2507.02798archive 2025-07-28

Miguel Espinosa, Chenhongyi Yang, Linus Ericsson, Steven McDonagh, Elliot J. Crowley

The performance of image segmentation models has historically been constrained by the high cost of collecting large-scale annotated data. The Segment Anything Model (SAM) alleviates this original problem through a promptable, semantics-agnostic, segmentation paradigm and yet still requires manual visual-prompts or complex domain-dependent prompt-generation rules to process a new image. Towards reducing this new burden, our work investigates the task of object segmentation when provided with, alternatively, only a small set of reference images. Our key insight is to leverage strong semantic priors, as learned by foundation models, to identify corresponding regions between a reference and a target image. We find that correspondences enable automatic generation of instance-level segmentation masks for downstream tasks and instantiate our ideas via a multi-stage, training-free method incorporating (1) memory bank construction; (2) representation aggregation and (3) semantic-aware feature matching. Our experiments show significant improvements on segmentation metrics, leading to state-of-the-art performance on COCO FSOD (36.8% nAP), PASCAL VOC Few-Shot (71.2% nAP50) and outperforming existing training-free approaches on the Cross-Domain FSOD benchmark (22.4% nAP).

PaperPDFCode

Code

miquel-espinosa/no-time-to-train officialmentioned on GitHubpytorch report
miquel-espinosa/samantics mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Cross-Domain Few-Shot Object DetectionFew-Shot Object DetectionImage SegmentationInstance SegmentationSegmentationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Cross-Domain Few-Shot Object Detection Artaxor Training-free(w/o FT) mAP 35.0 #8 of 16 Archive leaderboard report
Cross-Domain Few-Shot Object Detection Clipark1k Training-free(w/o FT) mAP 25.9 #6 of 10 Archive leaderboard report
Cross-Domain Few-Shot Object Detection DIOR Training-free(w/o FT) mAP 16.4 #11 of 15 Archive leaderboard report
Cross-Domain Few-Shot Object Detection DeepFish Training-free(w/o FT) mAP 29.6 #3 of 10 Archive leaderboard report
Cross-Domain Few-Shot Object Detection NEU-DET Training-free(w/o FT) mAP 5.5 #7 of 10 Archive leaderboard report
Cross-Domain Few-Shot Object Detection UODD Training-free(w/o FT) mAP 16.0 #7 of 16 Archive leaderboard report
Few-Shot Object Detection MS-COCO (1-shot) Training-free AP 26.5 #1 of 7 Archive leaderboard report
Few-Shot Object Detection MS-COCO (10-shot) Training-free AP 36.6 #1 of 33 Archive leaderboard report
Few-Shot Object Detection MS-COCO (30-shot) Training-free AP 36.8 #1 of 25 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

SET

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections