Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces XCT-SAM, a sequential parameter-efficient adaptation framework designed to enhance defect segmentation in additive manufacturing X-ray computed tomography (XCT) images. By employing Conv-LoRA adapters to fine-tune on an alloy-microstructure dataset before transferring to XCT data, XCT-SAM effectively bridges the significant domain gap and addresses the challenge of class imbalance. The method achieves superior performance over zero-shot SAM and other domain-adapted models, demonstrating its capability to maintain high accuracy in defect detection with minimal parameter adjustments.
XCT-SAM achieves state-of-the-art defect segmentation in XCT images by leveraging intermediate domain adaptation, outperforming traditional methods despite severe class imbalances and domain shifts.
Defect segmentation in additive manufacturing (AM) X-ray computed tomography (XCT) images remains challenging due to severe class imbalance and large distribution shifts across scan conditions. Although recent foundation models such as the Segment Anything Model (SAM) provide strong general-purpose segmentation priors, their natural-image pre-training transfers poorly to the AM XCT domain, where defects appear as subtle non-semantic microstructural anomalies. Moreover, adapting SAM to the AM domain is further limited by the large domain gap and scarcity of labeled real XCT data. We present XCT-SAM, a sequential parameter-efficient adaptation framework for AM XCT defect segmentation. Instead of adapting SAM directly from natural images to XCT data, we first fine-tune Conv-LoRA adapters on an alloy-microstructure dataset and subsequently transfer the adapted model to XCT images, progressively bridging the domain gap. Using Conv-LoRA adapters with rank r=2, the framework injects convolutional spatial inductive bias into SAM's backbone while training approximately 4.15M parameters and keeping over 99% of the model frozen. We evaluate XCT-SAM on out-of-distribution CycleGAN-XCT benchmarks and real-world NIST XCT scans. Across both settings, XCT-SAM consistently outperforms zero-shot SAM and other domain-adapted SAM baselines, achieving the best overall IoU and Dice scores. These results demonstrate the effectiveness of intermediate domain adaptation with parameter-efficient adapters for industrial XCT defect segmentation. The source code is publicly available at https://github.com/Mahedi-61/XCT-SAM.git