The paper introduces *Intelligent Scissors*, an interactive tool for image segmentation and composition. It addresses the challenges of fully automated segmentation and manual tracing by allowing users to extract objects from digital images quickly and accurately using simple mouse gestures. The tool uses a live-wire boundary detection method, which formulates discrete dynamic programming (DP) as a two-dimensional graph searching problem. This method provides mathematically optimal boundaries while reducing sensitivity to local noise. Additional features include *boundary cooling* for automatic freezing of unchanging segments and *on-the-fly training* for adapting to specific edge types. The extracted objects can be scaled, rotated, and composited using live-wire masks and *spatial frequency equivalencing*. The paper also discusses the background of image segmentation techniques, compares Intelligent Scissors with other methods like snakes, and presents results demonstrating the tool's effectiveness in various image types, including color images and medical imaging. The authors conclude by outlining future extensions and applications of Intelligent Scissors.The paper introduces *Intelligent Scissors*, an interactive tool for image segmentation and composition. It addresses the challenges of fully automated segmentation and manual tracing by allowing users to extract objects from digital images quickly and accurately using simple mouse gestures. The tool uses a live-wire boundary detection method, which formulates discrete dynamic programming (DP) as a two-dimensional graph searching problem. This method provides mathematically optimal boundaries while reducing sensitivity to local noise. Additional features include *boundary cooling* for automatic freezing of unchanging segments and *on-the-fly training* for adapting to specific edge types. The extracted objects can be scaled, rotated, and composited using live-wire masks and *spatial frequency equivalencing*. The paper also discusses the background of image segmentation techniques, compares Intelligent Scissors with other methods like snakes, and presents results demonstrating the tool's effectiveness in various image types, including color images and medical imaging. The authors conclude by outlining future extensions and applications of Intelligent Scissors.