mlsanityData to 14 Oct 2026

From Boundaries to Semantics: Prompt-Guided Multi-Task Learning for Petrographic Thin-section Segmentation

arXiv 2604.14805 ↗First posted 16 Apr 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Grain-edge segmentation (GES) and lithology semantic segmentation (LSS) are two pivotal tasks for quantifying rock fabric and composition. However, these two tasks are often treated separately, and the segmentation quality is implausible albeit expensive, time-consuming, and expert-annotated datasets have been used. Recently, foundation models, especially the Segment Anything Model (SAM), have demonstrated impressive robustness for boundary alignment. However, directly adapting SAM to joint GES and LSS is nontrivial due to 1) severe domain gap induced by extinction-dependent color variations and ultra-fine grain boundaries, and 2) lacking novel modules for joint learning on multi-angle petrographic image stacks. In this paper, we propose Petro-SAM, a novel two-stage, multi-task framework that can achieve high-quality joint GES and LSS on petrographic images. Specifically, based on SAM, we introduce a Merge Block to integrate seven polarized views, effectively solving the extinction issue. Moreover, we introduce multi-scale feature fusion and color-entropy priors to refine the detection.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 34 of its 55 references that are in the map. First posted on arXiv 16 Apr 2026.

Builds on

34 of 55 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  5. 2023Segment Anythingcited by 8,234
  6. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  7. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,940
  8. 2016Pyramid Scene Parsing Networkcited by 2,479
  9. 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,351
  10. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,161
  11. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
  12. 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,222
22 more
  1. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,135
  2. 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 803
  3. 2023LISA: Reasoning Segmentation via Large Language Modelcited by 776
  4. 2018CCNet: Criss-Cross Attention for Semantic Segmentationcited by 718
  5. 2015Holistically-Nested Edge Detectioncited by 592
  6. 2018End-to-End Multi-Task Learning with Attentioncited by 534
  7. 2018Boundary loss for highly unbalanced segmentationcited by 190
  8. 2018DLOW: Domain Flow for Adaptation and Generalizationcited by 177
  9. 2021Pixel Difference Networks for Efficient Edge Detectioncited by 168
  10. 2019Domain Adaptation for Structured Output via Discriminative Patch Representationscited by 150
  11. 2016Richer Convolutional Features for Edge Detectioncited by 143
  12. 2019Dense Extreme Inception Network: Towards a Robust CNN Model for Edge Detectioncited by 104
  13. 2019Bi-Directional Cascade Network for Perceptual Edge Detectioncited by 72
  14. 2014Fast Edge Detection Using Structured Forestscited by 65
  15. 2023SA-Med2D-20M Dataset: Segment Anything in 2D Medical Imaging with 20 Million maskscited by 55
  16. 2022Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularizationcited by 50
  17. 2024MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learningcited by 44
  18. 2023EdgeSAM: Prompt-In-the-Loop Distillation for SAMcited by 41
  19. 2017Latent Multi-task Architecture Learningcited by 29
  20. 2019Distance Map Loss Penalty Term for Semantic Segmentationcited by 27
  21. 2024SAUGE: Taming SAM for Uncertainty-Aligned Multi-Granularity Edge Detectioncited by 7
  22. 2025Prompt-Tuning SAM: From Generalist to Specialist with only 2048 Parameters and 16 Training Imagescited by 3

Built on by

Not cited yet.