In this paper, we study the salient object detection problem for images. We formulate this problem as a binary labeling task where we separate the salient object from the background. We propose a set of novel features, including multiscale contrast, center-surround histogram, and color spatial distribution, to describe a salient object locally, regionally, and globally. A conditional random field is learned to effectively combine these features for salient object detection. Further, we extend the proposed approach to detect a salient object from sequential images by introducing the dynamic salient features. We collected a large image database containing tens of thousands of carefully labeled images by multiple users and a video segment database, and conducted a set of experiments over them to demonstrate the effectiveness of the proposed approach.

/pdf/learning-to-detect-a-salient-object-19dnoso19v.pdf

Learning to Detect a Salient Object

Over the years, datasets and benchmarks have proven their fundamental importance in computer vision research, enabling targeted progress and objective comparisons in many fields. At the same time, legacy datasets may impend the evolution of a field due to saturated algorithm performance and the lack of contemporary, high quality data. In this work we present a new benchmark dataset and evaluation methodology for the area of video object segmentation. The dataset, named DAVIS (Densely Annotated VIdeo Segmentation), consists of fifty high quality, Full HD video sequences, spanning multiple occurrences of common video object segmentation challenges such as occlusions, motionblur and appearance changes. Each video is accompanied by densely annotated, pixel-accurate and per-frame ground truth segmentation. In addition, we provide a comprehensive analysis of several state-of-the-art segmentation approaches using three complementary metrics that measure the spatial extent of the segmentation, the accuracy of the silhouette contours and the temporal coherence. The results uncover strengths and weaknesses of current approaches, opening up promising directions for future works.

/pdf/a-benchmark-dataset-and-evaluation-methodology-for-video-4eb3ugu88k.pdf

A Benchmark Dataset and Evaluation Methodology for Video Object Segmentation

We present an efficient and scalable technique for spatiotemporal segmentation of long video sequences using a hierarchical graph-based algorithm. We begin by over-segmenting a volumetric video graph into space-time regions grouped by appearance. We then construct a “region graph” over the obtained segmentation and iteratively repeat this process over multiple levels to create a tree of spatio-temporal segmentations. This hierarchical approach generates high quality segmentations, which are temporally coherent with stable region boundaries, and allows subsequent applications to choose from varying levels of granularity. We further improve segmentation quality by using dense optical flow to guide temporal connections in the initial graph. We also propose two novel approaches to improve the scalability of our technique: (a) a parallel out-of-core algorithm that can process volumes much larger than an in-core algorithm, and (b) a clip-based processing algorithm that divides the video into overlapping clips in time, and segments them successively while enforcing consistency. We demonstrate hierarchical segmentations on video shots as long as 40 seconds, and even support a streaming mode for arbitrarily long videos, albeit without the ability to process them hierarchically.

/pdf/efficient-hierarchical-graph-based-video-segmentation-wchyj9kfja.pdf

Efficient hierarchical graph-based video segmentation

We present a technique for separating foreground objects from the background in a video. Our method is fast, fully automatic, and makes minimal assumptions about the video. This enables handling essentially unconstrained settings, including rapidly moving background, arbitrary object motion and appearance, and non-rigid deformations and articulations. In experiments on two datasets containing over 1400 video shots, our method outperforms a state-of-the-art background subtraction technique [4] as well as methods based on clustering point tracks [6, 18, 19]. Moreover, it performs comparably to recent video object segmentation methods based on object proposals [14, 16, 27], while being orders of magnitude faster.

/pdf/fast-object-segmentation-in-unconstrained-video-1z2edwzz4x.pdf

Fast Object Segmentation in Unconstrained Video

We present an approach to discover and segment foreground object(s) in video. Given an unannotated video sequence, the method first identifies object-like regions in any frame according to both static and dynamic cues. We then compute a series of binary partitions among those candidate “key-segments” to discover hypothesis groups with persistent appearance and motion. Finally, using each ranked hypothesis in turn, we estimate a pixel-level object labeling across all frames, where (a) the foreground likelihood depends on both the hypothesis's appearance as well as a novel localization prior based on partial shape matching, and (b) the background likelihood depends on cues pulled from the key-segments' (possibly diverse) surroundings observed across the sequence. Compared to existing methods, our approach automatically focuses on the persistent foreground regions of interest while resisting oversegmentation. We apply our method to challenging benchmark videos, and show competitive or better results than the state-of-the-art.

/pdf/key-segments-for-video-object-segmentation-1l1q5s5n4f.pdf

Key-segments for video object segmentation

Although tremendous success has been achieved for interactive object cutout in still images, accurately extracting dynamic objects in video remains a very challenging problem. Previous video cutout systems present two major limitations: (1) reliance on global statistics, thus lacking the ability to deal with complex and diverse scenes; and (2) treating segmentation as a global optimization, thus lacking a practical workflow that can guarantee the convergence of the systems to the desired results.We present Video SnapCut, a robust video object cutout system that significantly advances the state-of-the-art. In our system segmentation is achieved by the collaboration of a set of local classifiers, each adaptively integrating multiple local image features. We show how this segmentation paradigm naturally supports local user editing and propagates them across time. The object cutout system is completed with a novel coherent video matting technique. A comprehensive evaluation and comparison is presented, demonstrating the effectiveness of the proposed system at achieving high quality results, as well as the robustness of the system against various types of inputs.

Video SnapCut: robust video object cutout using localized classifiers

We introduce a novel approach to example-based stylization of portrait videos that preserves both the subject's identity and the visual richness of the input style exemplar. Unlike the current state-of-the-art based on neural style transfer [Selim et al. 2016], our method performs non-parametric texture synthesis that retains more of the local textural details of the artistic exemplar and does not suffer from image warping artifacts caused by aligning the style exemplar with the target face. Our method allows the creation of videos with less than full temporal coherence [Ruder et al. 2016]. By introducing a controllable amount of temporal dynamics, it more closely approximates the appearance of real hand-painted animation in which every frame was created independently. We demonstrate the practical utility of the proposed solution on a variety of style exemplars and target videos.

Example-based synthesis of stylized facial animations

Methods and apparatus, including computer program products, implementing and using techniques for generating a composite image from both two-dimensional (2D) image layers and three-dimensional (3D) image layers. In one embodiment, the image layers are arranged in a sequence in which a top 3D image layer is separated from a bottom 3D image layer by a 2D image layer and the image layers are rendered according to the sequence. In another embodiment, at least one of the image layers is associated with an adjustment layer, and rendering the composite image includes determining whether each current image layer has an associated adjustment layer with a dimensionality different from that of the current image layer, and if it does, applying the adjustment layer accordingly. In another embodiment, at least one of the image layers is associated with a track matte.

Compositing two-dimensional and three-dimensional image layers

Extracting temporally-coherent alpha mattes in video is an important but challenging problem in post-production Previous video matting systems are highly sensitive to initial conditions and image noise, thus cannot reliably produce stable alpha mattes without temporal jitter In this paper we propose an improved video matting system which contains two new components: (1) an accurate trimap propagation mechanism for setting up the initial matting conditions in a temporally-coherent way; and (2) a temporal matte filter which can improve the temporal coherence of the mattes while maintaining the matte structures on individual frames Experimental results show that compared with previous methods, the two new components lead to alpha mattes with better temporal coherence

Towards temporally-coherent video matting

Methods and apparatus for adaptive trimap propagation. Methods are described that allow a trimap to be propagated from one frame to the next in a temporally coherent way. A radius-based method propagates automatically computed local trimap radii from frame to frame. A mesh-based method employs pins on the binary segmentation boundary and a mesh generated for the unknown region; the pins are tracked from one frame to the next according to an optical flow technique, the mesh is deformed from one frame to the next according to the movement of the pins, and the adaptive trimap is then warped according to the deformed mesh. These methods can be used separately, or the first method can be used to propagate some regions of the adaptive trimap, and the second method can be used to propagate other regions of the adaptive trimap.

David P. Simons

Papers

Video SnapCut: robust video object cutout using localized classifiers

Example-based synthesis of stylized facial animations

Compositing two-dimensional and three-dimensional image layers

Towards temporally-coherent video matting

Adaptive Trimap Propagation for Video Matting