Filters








35,516 Hits in 2.9 sec

Self-supervised Video Object Segmentation [article]

Fangrui Zhu, Li Zhang, Yanwei Fu, Guodong Guo, Weidi Xie
2020 arXiv   pre-print
The objective of this paper is self-supervised representation learning, with the goal of solving semi-supervised video object segmentation (a.k.a. dense tracking).  ...  tasks; (iv) we demonstrate state-of-the-art results among the self-supervised approaches on DAVIS-2017 and YouTube-VOS, as well as surpassing most of methods trained with millions of manual segmentation  ...  Several recent studies [3] [4] [5] [6] [7] present promising results on self-supervised video object segmentation.  ... 
arXiv:2006.12480v1 fatcat:eeivxrmcrbdyfbcuynxmrn4xhm

Self-supervised Video Object Segmentation by Motion Grouping [article]

Charig Yang, Hala Lamdouar, Erika Lu, Andrew Zisserman, Weidi Xie
2021 arXiv   pre-print
In this paper, we work towards developing a computer vision system able to segment objects by exploiting motion cues, i.e. motion segmentation.  ...  the importance of motion cues, and the potential bias towards visual appearance in existing video segmentation models.  ...  video object segmentation (semi-supervised VOS), and unsupervised video object segmentation (unsupervised VOS).  ... 
arXiv:2104.07658v2 fatcat:mioxgxaekresznsbavtmpslhny

Self-Supervised Video Object Segmentation by Motion-Aware Mask Propagation [article]

Bo Miao, Mohammed Bennamoun, Yongsheng Gao, Ajmal Mian
2021 arXiv   pre-print
We propose a self-supervised spatio-temporal matching method, coined Motion-Aware Mask Propagation (MAMP), for video object segmentation.  ...  Moreover, MAMP performs at par with many supervised video object segmentation methods. Our code is available at: https://github.com/bo-miao/MAMP.  ...  Moreover, other similar looking non-target objects may confuse the model to segment incorrect objects. Semi-supervised VOS techniques fall into two categories: supervised and self-supervised.  ... 
arXiv:2107.12569v2 fatcat:escvlh6fsbclllgv34xtsd5paq

RPM-Net: Robust Pixel-Level Matching Networks for Self-Supervised Video Object Segmentation [article]

Youngeun Kim, Seokeon Choi, Hankyeol Lee, Taekyung Kim, Changick Kim
2019 arXiv   pre-print
Moreover, we significantly reduce the performance gap between self-supervised and fully-supervised video object segmentation (41.0% vs. 52.5% on DAVIS-2017 validation set)  ...  In this paper, we introduce a self-supervised approach for video object segmentation without human labeled data.Specifically, we present Robust Pixel-level Matching Net-works (RPM-Net), a novel deep architecture  ...  To this end, we introduce Robust Pixel-Level Matching Networks (RPM-Net) for self-supervised video object segmentation.  ... 
arXiv:1909.13247v2 fatcat:jnbj2ryfhfcqbgerpd54h5ilzm

Self-Supervised Video Object Segmentation via Cutout Prediction and Tagging [article]

Jyoti Kini and Fahad Shahbaz Khan and Salman Khan and Mubarak Shah
2022 arXiv   pre-print
We propose a novel self-supervised Video Object Segmentation (VOS) approach that strives to achieve better object-background discriminability for accurate object segmentation.  ...  Distinct from previous self-supervised VOS methods, our approach is based on a discriminative learning loss formulation that takes into account both object and background information to ensure object-background  ...  INTRODUCTION Video object segmentation (VOS) aims to segment an object of interest in a video, given its segmentation mask in the first frame.  ... 
arXiv:2204.10846v1 fatcat:ztskcslx6vg4bme3n6ja4li2xu

Self-supervised Object Tracking with Cycle-consistent Siamese Networks [article]

Weihao Yuan, Michael Yu Wang, Qifeng Chen
2020 arXiv   pre-print
The experiments on the VOT dataset for visual object tracking and on the DAVIS dataset for video object segmentation propagation show that our method outperforms prior approaches on both tasks.  ...  In this work, we exploit an end-to-end Siamese network in a cycle-consistent self-supervised framework for object tracking.  ...  Self-supervised Video Object Segmentation Propagation Training. For the video segmentation propagation task, we use the network with three branches.  ... 
arXiv:2008.00637v1 fatcat:trlfewfvgbbjpkhysiv5q4rusq

Self-supervised Segmentation by Grouping Optical-Flow [chapter]

Aravindh Mahendran, James Thewlis, Andrea Vedaldi
2019 Lecture Notes in Computer Science  
We propose to self-supervise a convolutional neural network operating on images using temporal information from videos.  ...  This learning by grouping approach is used as a pre-training as well as segmentation strategy.  ...  Therefore a segmentation with our objective will learn to segment objects or object-parts. We call this framework Self-Supervised Segmentation-CNN or S3-CNN .  ... 
doi:10.1007/978-3-030-11021-5_31 fatcat:337u756u6raf5no2fybbndq5nm

SPFTN: A Self-Paced Fine-Tuning Network for Segmenting Objects in Weakly Labelled Videos

Dingwen Zhang, Le Yang, Deyu Meng, Dong Xu, Junwei Han
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Object segmentation in weakly labelled videos is an interesting yet challenging task, which aims at learning to perform category-specific video object segmentation by only using video-level tags.  ...  To this end, we propose a novel self-paced fine-tuning network (SPFTN)-based framework, which could learn to explore the context information within the video frames and capture adequate object semantics  ...  The proposed self-paced fine-tuning network-based framework for object segmentation in weakly labelled videos.  ... 
doi:10.1109/cvpr.2017.567 dblp:conf/cvpr/ZhangYMXH17 fatcat:ixj26t5cmzgcbl27ushtb7nwni

The Emergence of Objectness: Learning Zero-Shot Segmentation from Videos [article]

Runtao Liu, Zhirong Wu, Stella X. Yu, Stephen Lin
2021 arXiv   pre-print
Our work is the first truly end-to-end zero-shot object segmentation from videos.  ...  Our model demonstrates the surprising emergence of objectness in the appearance pathway, surpassing prior works on zero-shot object segmentation from an image, moving object segmentation from a video with  ...  A popular objective for self-supervised learning from videos is view synthesis.  ... 
arXiv:2111.06394v1 fatcat:fujfxghw2vfdphl6dtyafsjo4i

Combining Self Training and Active Learning for Video Segmentation

Alireza Fathi, Maria Florina Balcan, Xiaofeng Ren, James M. Rehg
2011 Procedings of the British Machine Vision Conference 2011  
We show that video object segmentation can be naturally cast as a semi-supervised learning problem and be efficiently solved using harmonic functions.  ...  This work addresses the problem of segmenting an object of interest out of a video.  ...  Incremental self-training: We develop an iterative solution to semi-supervised video segmentation.  ... 
doi:10.5244/c.25.78 dblp:conf/bmvc/FathiBRR11 fatcat:persb76c35go3fdxhulee3gwny

Box Supervised Video Segmentation Proposal Network [article]

Tanveer Hannan, Rajat Koner, Jonathan Kobold, Matthias Schubert
2022 arXiv   pre-print
Video Object Segmentation (VOS) has been targeted by various fully-supervised and self-supervised approaches.  ...  In this work, we propose a box-supervised video object segmentation proposal network, which takes advantage of intrinsic video properties.  ...  Introduction Video Object Segmentation (VOS) primarily consists of two stages.  ... 
arXiv:2202.07025v2 fatcat:alq22qzyjvfhpi5eyxbt7zbcza

MAST: A Memory-Augmented Self-supervised Tracker [article]

Zihang Lai, Erika Lu, Weidi Xie
2020 arXiv   pre-print
Third, we benchmark on large-scale semi-supervised video object segmentation(aka. dense tracking), and propose a new metric: generalizability.  ...  We propose a dense tracking model trained on videos without any annotations that surpasses previous self-supervised methods on existing benchmarks by a significant margin (+15%), and achieves performance  ...  Youtube Video Object Segmentation We also evaluate the MAST model on the Youtube-VOS validation split (474 videos with 91 object categories).  ... 
arXiv:2002.07793v2 fatcat:hn6fof2ganfuldzbxkuvouckoq

Tracking Emerges by Colorizing Videos [chapter]

Carl Vondrick, Abhinav Shrivastava, Alireza Fathi, Sergio Guadarrama, Kevin Murphy
2018 Lecture Notes in Computer Science  
Moreover, our results suggest that failures to track are correlated with failures to colorize, indicating that advancing video colorization may further improve self-supervised visual tracking.  ...  We use large amounts of unlabeled video to learn models for visual tracking without manual human supervision.  ...  We also develop a self-supervised model, but our approach focuses on visual tracking in video for segmentation and human pose.  ... 
doi:10.1007/978-3-030-01261-8_24 fatcat:uhhkvzxgrjaqtnwdc7pxxb77fq

Tracking Emerges by Colorizing Videos [article]

Carl Vondrick, Abhinav Shrivastava, Alireza Fathi, Sergio Guadarrama, Kevin Murphy
2018 arXiv   pre-print
Moreover, our results suggest that failures to track are correlated with failures to colorize, indicating that advancing video colorization may further improve self-supervised visual tracking.  ...  We use large amounts of unlabeled video to learn models for visual tracking without manual human supervision.  ...  Example Video Segmentations: We show results from our self-supervised model on the task of video segmentation. Colors indicate different instances.  ... 
arXiv:1806.09594v2 fatcat:c5y74nmepnfvxiaknfcsuasut4

Self-Supervised Representation Learning for Visual Anomaly Detection [article]

Rabia Ali, Muhammad Umar Karim Khan, Chong Min Kyung
2020 arXiv   pre-print
While classification, object detection, and segmentation have been investigated with self-supervised learning, anomaly detection needs more attention.  ...  The feature representation obtained by self-supervision can be used in downstream tasks such as classification, object detection, segmentation, and anomaly detection.  ...  [41] use video colorization as a self-supervised learning problem. Wang and Gupta [42] propose a way of selfsupervised representation learning by tracking moving objects in videos.  ... 
arXiv:2006.09654v1 fatcat:otukxu2jufdefg2ktlymegghcy
« Previous Showing results 1 — 15 out of 35,516 results