Video Segmentation by Tracking Many Figure-Ground Segments

By Fuxin Li, Taeyoung Kim, Ahmad Humayun, David Tsai, James M. Rehg. IEEE International Conference on Computer Vision, 2013

589 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/203ea8ab1d9c48977be97e6caf3fdbcc84101354

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

Citations in the corpus

SegFlow: Joint Learning for Video Object Segmentation and Optical Flow (2017, methodology): “To incorporate higher level information such as objectness, object proposals are used to track object segments and generate consistent regions through the video [22, 23].”
Learning Video Object Segmentation from Static Images (2016, methodology): “Datasets We evaluate the proposed approach on three different video object segmentation datasets: DAVIS [30], YoutubeObjects [35], and SegTrack-v2 [24].”
Video Object Segmentation with Joint Re-identification and Attention-Aware Mask Propagation (2018, methodology): “To demonstrate the effectiveness and generalization ability of DyeNet, we evaluate our method on DAVIS 2016 [27], DAVIS 2017 [28], SegTrackv2 [19] and YouTubeObjects [29] datasets.”
Robust video segment proposals with painless occlusion handling (2015, methodology): “Post-processing using a conditional random field or composite statistical inference [31] could be used to further improve performance.”
DMM-Net: Differentiable Mask-Matching Network for Video Object Segmentation (2019, methodology): “On DAVIS 2017 [30], SegTrack v2 [20] and YouTube-VOS [45] datasets, our model achieves either state-of-the-art or comparable performance.”
MOTS: Multi-Object Tracking and Segmentation (2019, methodology): “In addition, the Segtrackv2 [28] dataset, FBMS [40] and an annotated subset of the YouTube-Objects dataset [46, 19] can be used to evaluate this task.”
Video Segmentation by Non-Local Consensus voting (2014, methodology): “We tested our algorithm on various datasets, ranging from benchmark evaluation datasets (SegTrack [21, 35]), on which we empirically compared results to others, to more difficult video sequences (taken from human action datasets like UCF101 [32] and ASLAN [19]), on which current state-of-the-art met”
Saliency-aware geodesic video object segmentation (2015, methodology): “We further carried out experiments on SegTrack v2 dataset [18] and 12 groups of videos randomly selected from Youtube Objects and compared our method with [32, 23, 15, 6] as well.”
Instance Embedding Transfer to Unsupervised Video Object Segmentation (2018, methodology): “Video segmentation datasets are smaller because they are more expensive to annotate [23, 30, 33].”
Video Salient Object Detection via Fully Convolutional Networks (2017, methodology): “In Table 1, we list the statistics of the ImageNet dataset and widely adopted video object segmentation datasets, including FBMS [10], SegTrackV2 [11], VSB100 [12] and DAVIS [13].”
Review of Visual Saliency Detection With Comprehensive Information (2018, methodology): “For video saliency detection, many datasets have been released, such as ViSal [98], MCL [99], UVSD [104], VOS [110], SegTrackV1 [119], SegTrackV2 [120], and DAVIS [121], as listed in Table VI.”
Video Segmentation via Object Flow (2016, methodology): “Note that, in contrast to unsupervised methods [18, 19], that rely on motion and object proposals and process the entire video offline in batch mode,”
Saliency-Aware Video Object Segmentation (2018, methodology): “Therefore, we report our performance on the IoU metric for the SegTrack [59], extended SegTrack [60], and FBMS [1] by computing the score for each frame and then averaging it over all frames.”
YouTube-VOS: A Large-Scale Video Object Segmentation Benchmark (2018, methodology): “In comparison to previous datasets [16,21,22,23,24,25], DAVIS has both higher-quality of video resolutions and annotations.”
Video Instance Segmentation Tracking With a Modified VAE Architecture (2020, methodology): “Popular datasets for the VOS task include DAVIS [50], Seg-TrackV2 [38] and YouTube-VOS [64].”
YouTube-VOS: Sequence-to-Sequence Video Object Segmentation (2018, methodology): “Scale JC [13] ST [26] YTO [21] FBMS [30] DAVIS [33] [34] YouTube-VOS (Ours)”