XCT-SAM: Sequential Parameter-Efficient Domain Adaptation of SAM for Industrial XCT Defect Segmentation
Md Mahedi Hasan, Md Mushfiqur Rahaman, Alan Pachkovskiy, Imtiaz Ahmed, Jeremy Dawson
実装難易度
Hard
推論・学習コスト
High
想定用途
セグメンテーション
概要
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…
何が新しいか
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.…
何に使えるか
セグメンテーション
実装情報
- Paper URL
- あり
- GitHub URL
- あり
実装チェックリスト
実装または配布ページ
OKコードまたはモデル配布ページから検証を始められます。
一次情報リンク
OKPaper / GitHub
検証しやすさ
要確認大規模モデル、生成、強化学習など再現コストが高い要素を含みます。
計算資源
要確認学習や高解像度推論ではGPUメモリと実行時間に注意が必要です。
ライセンス
未取得配布元のLICENSE、モデルカード、Paperの利用条件を確認してください。
商用利用
未取得研究利用限定、データセット由来制限、API規約の有無を確認してください。
自社データで試すなら
製造業・材料開発のExcel/CSVデータに落とし込むための最初の手順です。
- 1まずExcel/CSVの実験条件、組成、物性値を1行1実験の表に整理します。
- 2LightGBMやRandom Forestなどのベースラインを先に作り、この手法と比較します。
- 3評価指標はR2/RMSE、AUC、異常検知の再現率、実験回数削減率など、現場の意思決定に近いものを選びます。
- 4SHAPや特徴量重要度で、効いている因子が物理・化学・工程知識と矛盾しないか確認します。
実装難易度
Hard - 大規模モデル、生成、強化学習など再現コストが高い要素を含みます。
必要リソース
- GPU目安: High
- データセット: 論文・リポジトリ側の指定を確認してください。
- 学習要否: 再学習や評価環境の準備が必要になる可能性があります。
- 学習や高解像度推論ではGPUメモリと実行時間に注意が必要です。
実務で使う場合の注意点
- ライセンスと商用利用条件は、Paper / GitHub / Hugging Face の配布元で確認してください。
- 精度、再現性、計算コストはデータセットや評価条件に依存します。
- 個人情報や機密データを扱う場合は、入力データの保存先と外部API利用条件を確認してください。
関連記事
ManifoldFlow: SPD-Relaxed Stiefel Layers with Learnable Singular Spectrum
Orthogonal and Stiefel layers give neural weights exact spectral control, but they also impose a strong modeli
AI-For-Beginners — 12 Weeks, 24 Lessons, AI for All!
AIの入門コースを提供する。
segmentation_models.pytorch — Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.
セマンティックシーケンス分割モデルのライブラリです。
Advanced Image Generation: Negative Prompt Optimization and Latent Classifier Guidance
画像生成を改善するための新しいシステムが提案されます。このシステムは、負の提示 (negative prompt) の最適化と潜在空間における分類器 (latent classifier) の導導導導導導導導導導導導導導