Intelligent scissors for image composition

By Eric N. Mortensen, William A. Barrett. 1995

We present a new, interactive tool called Intelligent Scissors which we use for image segmentation and composition. Fully automated segmentation is an unsolved problem, while manual tracing is inaccurate and laboriously unacceptable. However, Intelligent Scissors allow objects within digital images to be extracted quickly and accurately using simple gesture motions with a mouse. When the gestured mouse position comes in proximity to an object edge, a live-wire boundary “snaps” to, and wraps around the object of interest. Live-wire boundary detection formulates discrete dynamic programming (DP) as a two-dimensional graph searching problem. DP provides mathematically optimal boundaries while greatly reducing sensitivity to local noise or other intervening structures. Robustness is further enhanced with on-the-fly training which causes the boundary to adhere to the specific type of edge currently being followed, rather than simply the strongest edge in the neighborhood. Boundary cooling automatically freezes unchanging segments and automates input of additional seed points. Cooling also allows the user to be much more free with the gesture path, thereby increasing the efficiency and finesse with which boundaries can be extracted. Extracted objects can be scaled, rotated, and composited using live-wire masks and spatial frequency equivalencing. Frequency equivalencing is performed by applying a Butterworth filter which matches the lowest frequency spectra to all other image components. Intelligent Scissors allow creation of convincing compositions from existing images while dramatically increasing the speed and precision with which objects can be extracted.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Human-Computer Interaction

Citations in the corpus

Deep Interactive Object Selection (2016, neutral): “There are many interactive segmentation approaches proposed over the past decades, such as contour-based methods [19, 10] and bounding box methods [21].”
User-Friendly Interactive Image Segmentation Through Unified Combinatorial User Inputs (2010, methodology): “The details about how to compute , , and can be found in [18].”
Geodesic graph cut for interactive image segmentation (2010, methodology): “, [3, 10, 13, 21, 30]) to loosely marking parts of the desired object and/or background (e.”
Predicting Sufficient Annotation Strength for Interactive Foreground Segmentation (2013, unknown): “to efficiently identify the foreground region [8, 16, 2, 19, 12, 7, 1].”
Error-Tolerant Scribbles Based Interactive Image Segmentation (2014, background): “The boundary-based interactive segmentation such as intelligent scissors [9] requires the user to trace the whole boundary of the object, which is usually time-consuming and tedious for users.”
An ultra-fast user-steered image segmentation paradigm: live wire on the fly (2000, methodology): “These properties set our approach apart from previous optimum boundary detection techniques including other live-wire methods [7], [18], [19] and active contour approaches.”
Fast Interactive Object Annotation With Curve-GCN (2019, methodology): “Intelligent Scissors [25] is a technique that allows the user to place “seeds” along the boundary and finds the minimal cost contour starting from the last seed up to the mouse cursor, by tracing along the object’s boundary.”
Lazy snapping (2004, background): “Examples include intelligent scissor [Mortensen and Barrett 1995; Mortensen and Barrett 1999], image snapping [Gleicher 1995] and Jetstream [Perez and Blake 2001].”
Semantic Object Selection (2014, background): “With Intelligent Scissors [18], the user traces the boundary of the object.”
GrabCut: Interactive Foreground Extraction Using Iterated Graph Cuts (2023, background): “Intelligent Scissors (a.k.a. Live Wire or Magnetic Lasso cost contour(cid:148) by roughly the object’s boundary with the mouse.”
Video SnapCut: robust video object cutout using localized classifiers (2009, background): “Among these, shape should play an important role for maintaining a temporallycoherent recognition; and (2) Multiple cues should be evaluated and integrated both locally and globally, rather than just globally, to maximize their discriminant powers.”
Interactive Image Segmentation with Latent Diversity (2018, background): “Research on interactive segmentation dates back decades, with early work focusing on boundary tracing techniques [18, 32].”
PhraseClick: Toward Achieving Flexible Interactive Segmentation by Phrase and Click (2020, methodology): “Early methods rely on low-level features, such as color similarity or boundary properties [26,45].”
Beyond pixels: A comprehensive survey from bottom-up to semantic image segmentation and cosegmentation (2015, methodology): “For instance, the Live-Wire / Intelligent Scissors method [73], [75] starts contour evolving by building a weighted graph on the image, where each node in the graph corresponds to a pixel, and directed edges are formed around pixels with their closest four neighbors or eight neighbors.”
Regional Interactive Image Segmentation Networks (2017, unknown): “Early interactive image segmentation methods, such as the parametric active contour model [14] and intelligent scissors [20] mainly consider boundary properties when performing segmentation, thus performing poorly at weak edges.”