Interactive Video Segmentation via Local Appearance Model

Sun, Chong; Lu, Huchuan

doi:10.1109/tcsvt.2016.2543038

Cited by 8 publications

(4 citation statements)

References 45 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One such a video segmentation method is focused on the regions within frames -e.g. to generate a binary mask for a given target object in each frame [25]. While some of these masking-video segmentation methodologies can be employed to measure the consistency or presence of a target in a frame, which can be evaluated to detect a possible shot boundary, it generally comes at a price as it is computationally expensive.…”

Section: Literature Overviewmentioning

confidence: 99%

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Klerk¹,

Venter²,

Hoffman³

2018

SAIEE Afr. Res. J.

View full text Add to dashboard Cite

Shot boundary detection is an integral part of multimedia, be it video management or video processing. Multiple boundary detection techniques have been developed throughout the years, but are only applicable to very specific instances. The Jensen-Shannon divergence (JSD) is one such a technique that can be implemented to detect the shot boundaries in digital videos. This paper investigates the use of the JSD algorithm to detect shot boundaries in streaming media applications. Furthermore, the effects of the various parameters used by the JSD technique, on the accuracy of the detected boundaries, are quantified by the recall and precision metrics all the while keeping track of how they affect the execution time.

show abstract

Section: Literature Overviewmentioning

confidence: 99%

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Klerk¹,

Venter²,

Hoffman³

2018

SAIEE Afr. Res. J.

View full text Add to dashboard Cite

show abstract

“…Most recent offline approaches are increasingly using complex data-driven models that rely on the availability of video segmentation datasets, which the recent benchmark of Perazzi et al [38] points out the scarcity of, as well as their limited size and variations. On the other side, interactive methods [5,39,47,32,41] rely on complex hand-crafted features and cues. This work leverages the diversity of propagation methods combined with a crowd workforce to implicitly select the best features and propagation strategies.…”

Section: Video Segmentation Propagationmentioning

confidence: 99%

Crowd-Guided Ensembles

Kaspar

Patterson

Kim

et al. 2018

Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

Figure 1. An illustration of our two proposed crowd-guided ensemble methods. Left: Our segmentation ensemble combines the results of multiple crowd workers through the guidance of an oracle reviewer. Right: Our propagation ensemble gathers the information about where multiple distinct algorithms fail from the accumulated scribbles of crowd workers and merges it into the result that incorporates the best of each algorithm.

show abstract

“…Naive methods use optical flow to model the frame to frame propagation [5] but often suffer from occlusion and reappearance of objects. To handle this problem, more sophisticated appearance models have been proposed, such as local shape models in [6], similarity graph in [7] and patch cross-correlation in [8]. Unstructured classifiers, which predict pixel labels independently without neighborhood constraints, mostly use Random Decision Forests [9] and CRFs to get the unaries.…”

Section: Introductionmentioning

confidence: 99%

Automatic Annotation of Airborne Images by Label Propagation Based on a Bayesian-CRF Model

et al. 2019

View full text Add to dashboard Cite

The tremendous advances in deep neural networks have demonstrated the superiority of deep learning techniques for applications such as object recognition or image classification. Nevertheless, deep learning-based methods usually require a large amount of training data, which mainly comes from manual annotation and is quite labor-intensive. In order to reduce the amount of manual work required for generating enough training data, we hereby propose to leverage existing labeled data to generate image annotations automatically. Specifically, the pixel labels are firstly transferred from one image modality to another image modality via geometric transformation to create initial image annotations, and then additional information (e.g., height measurements) is incorporated for Bayesian inference to update the labeling beliefs. Finally, the updated label assignments are optimized with a fully connected conditional random field (CRF), yielding refined labeling for all pixels in the image. The proposed approach is tested on two different scenarios, i.e., (1) label propagation from annotated aerial imagery to unmanned aerial vehicle (UAV) imagery and (2) label propagation from map database to aerial imagery. In each scenario, the refined image labels are used as pseudo-ground truth data for training a convolutional neural network (CNN). Results demonstrate that our model is able to produce accurate label assignments even around complex object boundaries; besides, the generated image labels can be effectively leveraged for training CNNs and achieve comparable classification accuracy as manual image annotations, more specifically, the per-class classification accuracy of the networks trained by the manual image annotations and the generated image labels have a difference within ± 5 % .

show abstract

Interactive Video Segmentation via Local Appearance Model

Cited by 8 publications

References 45 publications

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Crowd-Guided Ensembles

Automatic Annotation of Airborne Images by Label Propagation Based on a Bayesian-CRF Model

Contact Info

Product

Resources

About