Video Segmentation via Object Flow

By Yi-Hsuan Tsai, Ming-Hsuan Yang, Michael J. Black. Computer Vision and Pattern Recognition, 2016

358 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/19351b059b2fabafd885322d26a39ed469265654

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

Saliency-Aware Video Object Segmentation (2018, methodology): “Table 2 presents the IoU scores of our method and [7], [9], [22], [25], [27], [29], [63], [64], [65], [66] on the SegTrack and the extended SegTrack.”
Fast Video Object Segmentation by Reference-Guided Mask Propagation (2018, background): “The results for OFL [38] and OSVOS [5] are directly copied from [18].”
Video Object Segmentation Using Space-Time Memory Networks (2019, methodology): “Therefore, we put these two frames into the memory by default.”
See More, Know More: Unsupervised Video Object Segmentation With Co-Attention Siamese Networks (2019, methodology): “Later, inspired by the success of deep learning, several methods [16, 54, 43] began to approach UVOS using deep learning features.”
FusionSeg: Learning to Combine Motion and Appearance for Fully Automatic Segmentation of Generic Objects in Videos (2017, methodology): “Human-guided methods Semi-supervised label propagation methods accept human input on a subset of frames, then propagate it to the remaining frames [40, 47, 1, 11, 50, 19, 53, 37, 31, 48].”
A Generative Appearance Model for End-To-End Video Object Segmentation (2018, methodology): “Some methods rely on optical flow coupled with refinement [15, 29].”
Data Augmentation Using Learned Transformations for One-Shot Medical Image Segmentation (2019, background): “Few-shot video segmentation is frequently implemented by aligning objects in each frame to a labeled reference frame [37, 75].”
Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks (2019, background): “, color, optical flow) [43, 12, 59, 20] and certain heuristic assumptions related to the foreground (i.”
Lucid Data Dreaming for Object Tracking (2017, methodology): “Mask propagation: Appearance similarity and motion smoothness across time is used to propagate the first frame annotation across the video [33] [60] [55].”
MASK-RL: Multiagent Video Object Segmentation Framework Through Reinforcement Learning (2020, background): “We then compare our model with user clicks against both semisupervised state-of-the-art approaches, namely, [5], [28], [35], [42], and weakly supervised ones, i.”
Video Object Segmentation without Temporal Information (2017, methodology): “We compare against a large body of very recent semi-supervised state-of-the-art techniques (OnAVOS [65], MSK [46], CTN [29], VPN [27], OFL [63], BVS [43], and FCP [48]) using the pre-computed results provided by the respective authors.”
DMM-Net: Differentiable Mask-Matching Network for Video Object Segmentation (2019, neutral): “4 OFL [36] 4 67.”
SegFlow: Joint Learning for Video Object Segmentation and Optical Flow (2017, methodology): “Compared to other methods such as OFL (30 seconds per frame for optical flow generation and 90 seconds per frame for optimization), MSK (12 seconds per frame) and OSVOS (more than 10 seconds per frame at its best performance), our method is faster and can output an additional result of optical flow.”
Learning Video Object Segmentation from Static Images (2016, methodology): “ObjFlow [44] builds a graph over pixels and superpixels, uses convnet based appearance terms, and interleaves labeling with optical ﬂow estimation.”
Instance Embedding Transfer to Unsupervised Video Object Segmentation (2018, methodology): “Early foreground segmentation methods often used Gaussian Mixture Models and Graph Cut [27, 38], but more recent work uses convolutional neural networks (CNN) to identify foreground pixels based on saliency, edges, and/or motion [36, 37, 16].”
Fast Online Object Tracking and Segmentation: A Unifying Approach (2018, methodology): “This result is achieved with a simple bounding box initialisation (as opposed to a mask) and without adopting costly techniques often used by VOS approaches such as fine-tuning [39, 45, 2, 60], data augmentation [25, 33] and optical flow [57, 2, 45, 33, 10].”
Online Model Distillation for Efficient Video Inference (2018, methodology): “JITNet A performs similarly to OFL [45], a flow-based approach for video object segmentation, while JITNet B, using OSVOS-S predictions, performs comparably to OSVOS, with significantly lower runtime cost.”