Pith. sign in

REVIEW 2 cited by

Active Label Correction for Semantic Segmentation with Foundation Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.10820 v2 pith:XLUPDKUU submitted 2024-03-16 cs.CV

Active Label Correction for Semantic Segmentation with Foundation Models

classification cs.CV
keywords correctionlabelmodelsactivedatasetsfoundationlabelspascal
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Training and validating models for semantic segmentation require datasets with pixel-wise annotations, which are notoriously labor-intensive. Although useful priors such as foundation models or crowdsourced datasets are available, they are error-prone. We hence propose an effective framework of active label correction (ALC) based on a design of correction query to rectify pseudo labels of pixels, which in turn is more annotator-friendly than the standard one inquiring to classify a pixel directly according to our theoretical analysis and user study. Specifically, leveraging foundation models providing useful zero-shot predictions on pseudo labels and superpixels, our method comprises two key techniques: (i) an annotator-friendly design of correction query with the pseudo labels, and (ii) an acquisition function looking ahead label expansions based on the superpixels. Experimental results on PASCAL, Cityscapes, and Kvasir-SEG datasets demonstrate the effectiveness of our ALC framework, outperforming prior methods for active semantic segmentation and label correction. Notably, utilizing our method, we obtained a revised dataset of PASCAL by rectifying errors in 2.6 million pixels in PASCAL dataset.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. iSAGE: A Human-in-the-Loop Framework for Remote Sensing Semantic Segmentation via Sparse Point Supervision

    cs.CV 2026-06 unverdicted novelty 7.0

    iSAGE achieves near-dense mIoU performance in remote sensing semantic segmentation using iterative expert clicks on confident model errors with an error-weighted loss, using only 0.011-0.04% of pixels.

  2. BACON: Budgeted Human Calibration for Modeling and Evaluation with Multiple AI Judges

    cs.LG 2026-06 conditional novelty 5.0

    BACON calibrates multiple AI judges against a small human-labeled sample, then uses cross-fitted outcome models and augmented estimating equations to produce calibrated summary estimates and item-level surrogate scores.