Tracking-based semi-supervised learning

Teichman, Alex; Thrun, Sebastian

doi:10.1177/0278364912442751

Cited by 53 publications

(33 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…object candidates) from video or streams of sensory recordings in general. Teichman et al [45] propose a method for tracking-based semi-supervised learning by mining Li-DAR streams, captured from a vehicle. Similarly, [28], [29] propose tracking-based semi-supervised learning based on video.…”

Section: Related Workmentioning

confidence: 99%

Large-Scale Object Mining for Object Discovery from Unlabeled Video

Ošep

Voigtlaender

Luiten

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

This paper addresses the problem of object discovery from unlabeled driving videos captured in a realistic automotive setting. Identifying recurring object categories in such raw video streams is a very challenging problem. Not only do object candidates first have to be localized in the input images, but many interesting object categories occur relatively infrequently. Object discovery will therefore have to deal with the difficulties of operating in the long tail of the object distribution. We demonstrate the feasibility of performing fully automatic object discovery in such a setting by mining object tracks using a generic object tracker. In order to facilitate further research in object discovery, we release a collection of more than 360,000 automatically mined object tracks from 10+ hours of video data (560,000 frames). We use this dataset to evaluate the suitability of different feature representations and clustering strategies for object discovery.

show abstract

Section: Related Workmentioning

confidence: 99%

Large-Scale Object Mining for Object Discovery from Unlabeled Video

Ošep

Voigtlaender

Luiten

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…In the meanwhile, we classify each of the obstacle clusters based on a multiboost classifier proposed in [18], which results in cars, bicycles, and pedestrians. Kalman filters are then created for each object for filtering and predicting their movements in adjacent frames (Fig.…”

Section: Perception Under Uncertaintymentioning

confidence: 99%

TiEV: The Tongji Intelligent Electric Vehicle in the Intelligent Vehicle Future Challenge of China

Zhao

Wang

Yan

et al. 2018

2018 21st International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

TiEV is an autonomous driving platform implemented by the Tongji University of China. The vehicle is driveby-wire and is fully powered by electricity. We devised the software system of TiEV from scratch, which is capable of driving the vehicle autonomously in urban paths as well as on fast express roads. We describe our whole system, especially novel modules of probabilistic perception fusion, large-scale mapping and updating, the 1 st and the 2 nd planning and the overall safety concern. TiEV finished 2016, and 2017 Intelligent Vehicle Future Challenge of China held at Changshu. We show our experiences on the development of autonomous vehicles and discuss the future trends.

show abstract

“…The framework is able to track various objects in limited drifting environments. The classification of objects that have been segmented and tracked without the use of a class-specific tracker has been addressed with an SSL algorithm in Teichman and Thrun (2011). Given only three hand-labelled training examples of each class, the algorithm can perform comparably to equivalent fully-supervised methods, but it requires full-length tracks (it is therefore an off-line process) generated by a perfect tracker (each stream represents a single object), which would be challenging for real applications, where multiple streams are available simultaneously.…”

Section: Literature Reviewmentioning

confidence: 99%

Learning from evolving video streams in a multi-camera scenario

2015

View full text Add to dashboard Cite

Nowadays, video surveillance systems are taking the first steps toward automation, in order to ease the burden on human resources as well as to avoid human error. As the underlying data distribution and the number of concepts change over time, the conventional learning algorithms fail to provide reliable solutions for this setting. In this paper, we formalize a learning concept suitable for multi-camera video surveillance and propose a learning methodology adapted to that new paradigm. The proposed framework resorts to the universal background model to robustly learn individual object models from small samples and to more effectively detect novel classes. The individual models are incrementally updated in an ensemble-based approach, with older models being progressively forgotten. The framework is designed to detect and label new concepts automatically. The system is also designed to exploit active learning strategies, in order to interact wisely with operator, requesting assistance in the most ambiguous to classify observations. The experimental results obtained both on real and synthetic data sets verify the usefulness of the proposed approach.

show abstract

Tracking-based semi-supervised learning

Cited by 53 publications

References 31 publications

Large-Scale Object Mining for Object Discovery from Unlabeled Video

Large-Scale Object Mining for Object Discovery from Unlabeled Video

TiEV: The Tongji Intelligent Electric Vehicle in the Intelligent Vehicle Future Challenge of China

Learning from evolving video streams in a multi-camera scenario

Contact Info

Product

Resources

About