Learning to find occlusion regions

Humayun, Ahmad; Aodha, Oisin Mac; Brostow, Gabriel J.

doi:10.1109/cvpr.2011.5995517

Cited by 49 publications

(50 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another set of algorithms infer occlusion boundaries (Stein and Hebert 2009;He and Yuille 2010) and occluded regions (Humayun et al 2011) training a learning based detector using appearance, motion and depth features. The accuracy of these methods largely depends on the performance of the underlying feature detectors.…”

Section: Prior Related Workmentioning

confidence: 99%

Sparse Occlusion Detection with Optical Flow

2011

View full text Add to dashboard Cite

We tackle the problem of detecting occluded regions in a video stream. Under assumptions of Lambertian reflection and static illumination, the task can be posed as a variational optimization problem, and its solution approximated using convex minimization. We describe efficient numerical schemes that reach the global optimum of the relaxed cost functional, for any number of independently moving objects, and any number of occlusion layers. We test the proposed algorithm on benchmark datasets, expanded to enable evaluation of occlusion detection performance, in addition to optical flow.

show abstract

Section: Prior Related Workmentioning

confidence: 99%

Sparse Occlusion Detection with Optical Flow

2011

View full text Add to dashboard Cite

show abstract

“…Detecting occlusion boundaries is a well studied problem, due to its usefulness in understanding the depth, motion and context of the scene [22,12]. Fleet et al [6] gave a Bayesian formulation where boundaries resulted from distinguishing local image motion.…”

Section: Related Researchmentioning

confidence: 99%

“…Therefore, pixels advected from reference frame I t by estimated flow F t→t+1 should correspond to the next frame I t+1 . This assumption breaks down at occlusion boundaries, hence high photoconsistency residual should be indicative of such boundaries [12,13]. Residual photo-consistency feature F PC is computed as…”

Section: Features For Occlusion Boundary Predictionmentioning

confidence: 99%

Finding Temporally Consistent Occlusion Boundaries in Videos Using Geometric Context

Raza

Humayun²,

Essa³

et al. 2015

2015 IEEE Winter Conference on Applications of Computer Vision

Self Cite

View full text Add to dashboard Cite

We present an algorithm for finding temporally consistent occlusion boundaries in videos to support segmentation of dynamic scenes. We learn occlusion boundaries in a pairwise Markov random field (MRF) framework. We first estimate the probability of an spatio-temporal edge being an occlusion boundary by using appearance, flow, and geometric features. Next, we enforce occlusion boundary continuity in a MRF model by learning pairwise occlusion probabilities using a random forest. Then, we temporally smooth boundaries to remove temporal inconsistencies in occlusion boundary estimation. Our proposed framework provides an efficient approach for finding temporally consistent occlusion boundaries in video by utilizing causality, redundancy in videos, and semantic layout of the scene. We have developed a dataset with fully annotated ground-truth occlusion boundaries of over 30 videos (∼5000 frames). This dataset is used to evaluate temporal occlusion boundaries and provides a much needed baseline for future studies. We perform experiments to demonstrate the role of scene layout, and temporal information for occlusion reasoning in dynamic scenes.

show abstract

“…Specifically, we compute the difference in color, geometric context, and motion features for the regions on both sides of an edgelet. In addition, we compute the flow-consistency features along each edgelet boundary [9]. We use standard pairwise potential term over edgelet occlusion prediction.…”

Section: Occlusion Boundariesmentioning

confidence: 99%

Depth Extraction from Videos Using Geometric Context and Occlusion Boundaries

Raza¹,

Javed²,

Das³

et al. 2014

Proceedings of the British Machine Vision Conference 2014

View full text Add to dashboard Cite

We present an algorithm to estimate depth in dynamic video scenes. We propose to learn and infer depth in videos from appearance, motion, occlusion boundaries, and geometric context of the scene. Using our method, depth can be estimated from unconstrained videos with no requirement of camera pose estimation, and with significant background/foreground motions. We start by decomposing a video into spatio-temporal regions. For each spatio-temporal region, we learn the relationship of depth to visual appearance, motion, and geometric classes. Then we infer the depth information of new scenes using piecewise planar parametrization estimated within a Markov random field (MRF) framework by combining appearance to depth learned mappings and occlusion boundary guided smoothness constraints. Subsequently, we perform temporal smoothing to obtain temporally consistent depth maps. To evaluate our depth estimation algorithm, we provide a novel dataset with ground truth depth for outdoor video scenes. We present a thorough evaluation of our algorithm on our new dataset and the publicly available Make3d static image dataset.

show abstract

Learning to find occlusion regions

Cited by 49 publications

References 47 publications

Sparse Occlusion Detection with Optical Flow

Sparse Occlusion Detection with Optical Flow

Finding Temporally Consistent Occlusion Boundaries in Videos Using Geometric Context

Depth Extraction from Videos Using Geometric Context and Occlusion Boundaries

Contact Info

Product

Resources

About