Ensemble Knowledge Transfer for Semantic Segmentation

Nigam, Ishan; Huang, Chen; Ramanan, Deva

doi:10.1109/wacv.2018.00168

Cited by 76 publications

(33 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Aeroscapes dataset is an aerial semantic segmentation benchmark dataset that was collected using a DJI Phantom 3 fleet of drones between the altitudes of 5 and 50 m [78]. The dataset consisted of 3269 720p non-sequential images and their respective 11 class annotations.…”

Section: Vehicle Detection and Segmentation Datasetmentioning

confidence: 99%

“…These constraints depleted an already scarce pool of candidate datasets, reducing it to few multi-class segmentation datasets. For object detection, we used the Aeroscapes dataset [78] and collected additional UAV data of vehicles using a fixed wing UAV for further testing purposes. Vehicle tracking and speed estimation were assessed using the Karlsruhe Institute for Technology Aerial Image Sequences (KIT AIS) dataset [79].…”

mentioning

confidence: 99%

See 1 more Smart Citation

MultEYE: Monitoring System for Real-Time Vehicle Detection, Tracking and Speed Estimation from UAV Imagery on Edge-Computing Platforms

2021

View full text Add to dashboard Cite

We present MultEYE, a traffic monitoring system that can detect, track, and estimate the velocity of vehicles in a sequence of aerial images. The presented solution has been optimized to execute these tasks in real-time on an embedded computer installed on an Unmanned Aerial Vehicle (UAV). In order to overcome the limitation of existing object detection architectures related to accuracy and computational overhead, a multi-task learning methodology was employed by adding a segmentation head to an object detector backbone resulting in the MultEYE object detection architecture. On a custom dataset, it achieved 4.8% higher mean Average Precision (mAP) score, while being 91.4% faster than the state-of-the-art model and while being able to generalize to different real-world traffic scenes. Dedicated object tracking and speed estimation algorithms have been then optimized to track reliably objects from an UAV with limited computational effort. Different strategies to combine object detection, tracking, and speed estimation are discussed, too. From our experiments, the optimized detector runs at an average frame-rate of up to 29 frames per second (FPS) on frame resolution 512 × 320 on a Nvidia Xavier NX board, while the optimally combined detector, tracker and speed estimator pipeline achieves speeds of up to 33 FPS on an image of resolution 3072 × 1728. To our knowledge, the MultEYE system is one of the first traffic monitoring systems that was specifically designed and optimized for an UAV platform under real-world constraints.

show abstract

Section: Vehicle Detection and Segmentation Datasetmentioning

confidence: 99%

mentioning

confidence: 99%

MultEYE: Monitoring System for Real-Time Vehicle Detection, Tracking and Speed Estimation from UAV Imagery on Edge-Computing Platforms

2021

View full text Add to dashboard Cite

show abstract

“…Direct interaction of drones can be implemented in the following positions: Thanks to the cameras, drones will be able to transmit images to the staff monitors who need to assess the environment. However, these cameras can also be used for positioning on terrain and overcoming visible obstacles (Nigam et al, 2018). Cameras may be also used for direct positioning among other drones.…”

Section: Interaction Of Drones In Its Suit Modelmentioning

confidence: 99%

Drone Hardware And Software Development And Testing Quality Management

Plotnikov¹,

Skvoroda²

2020

European Proceedings of Social and Behavioural Sciences

View full text Add to dashboard Cite

show abstract

“…Semantic segmentation plays a very important role in understanding and interpreting a scene, we can use semantic segmentation in space research for detecting inter-galactic activities and enhance their knowledge. [20] considers an ensemble model incorporating knowledge transfer based on drones for the semantic segmentation of aerial images. The Passive Aggressive(PA) algorithm is popular for the online update of models.…”

Section: Related Workmentioning

confidence: 99%

Online Transfer Learning and Organic Computing for Deep Space Research and Astronomy

Natarajan

2019

SSRN Journal

View full text Add to dashboard Cite

Deep space exploration is the pillars within the field of outer space analysis and physical science. The amount of knowledge from numerous space vehicle and satellites orbiting the world of study are increasing day by day. This information collected from numerous experiences of the advanced space missions is huge. These information helps us to enhance current space knowledge and the experiences can be converted and transformed into segregated knowledge which helps us to explore and understand the realms of the deep space.. Online Transfer Learning (OTL) is a machine learning concept in which the knowledge gets transferred between the source domain and target domain in real time, in order to help train a classifier of the target domain. Online transfer learning can be an efficient method for transferring experiences and data gained from the space analysis data to a new learning task and can also routinely update the knowledge as the task evolves.

show abstract

Ensemble Knowledge Transfer for Semantic Segmentation

Cited by 76 publications

References 31 publications

MultEYE: Monitoring System for Real-Time Vehicle Detection, Tracking and Speed Estimation from UAV Imagery on Edge-Computing Platforms

MultEYE: Monitoring System for Real-Time Vehicle Detection, Tracking and Speed Estimation from UAV Imagery on Edge-Computing Platforms

Drone Hardware And Software Development And Testing Quality Management

Online Transfer Learning and Organic Computing for Deep Space Research and Astronomy

Contact Info

Product

Resources

About