End-to-end active object tracking and its real-world deployment via reinforcement learning
We study active object tracking, where a tracker takes visual observations (ie, frame
sequences) as input and produces the corresponding camera control signals as output (eg …
sequences) as input and produces the corresponding camera control signals as output (eg …
End-to-end active object tracking via reinforcement learning
We study active object tracking, where a tracker takes as input the visual observation (ie
frame sequence) and produces the camera control signal (eg, move forward, turn left, etc) …
frame sequence) and produces the camera control signal (eg, move forward, turn left, etc) …
Ad-vat+: An asymmetric dueling mechanism for learning and understanding visual active tracking
Visual Active Tracking (VAT) aims at following a target object by autonomously controlling
the motion system of a tracker given visual observations. To learn a robust tracker for VAT, in …
the motion system of a tracker given visual observations. To learn a robust tracker for VAT, in …
AD-VAT: An asymmetric dueling mechanism for learning visual active tracking
Visual Active Tracking (VAT) aims at following a target object by autonomously controlling
the motion system of a tracker given visual observations. Previous work has shown that the …
the motion system of a tracker given visual observations. Previous work has shown that the …
Coordinate-aligned multi-camera collaboration for active multi-object tracking
Abstract Active Multi-Object Tracking (AMOT) is a task where cameras are controlled by a
centralized system to adjust their poses automatically and collaboratively so as to maximize …
centralized system to adjust their poses automatically and collaboratively so as to maximize …
An improved method based on deep reinforcement learning for target searching
XL Wei, XL Huang, T Lu… - 2019 4th international …, 2019 - ieeexplore.ieee.org
Unmanned Aerial Vehicle (UAV), due to their high mobility and the ability to cover areas of
different heights and locations at relatively low cost, are increasingly used for disaster …
different heights and locations at relatively low cost, are increasingly used for disaster …
Deep reinforcement learning for autonomous search and rescue
Unmanned Aerial Vehicles (UAVs) are becoming more prevalent, more capable, and less
expensive every day. Advances in battery life and electronic sensors have spurred the …
expensive every day. Advances in battery life and electronic sensors have spurred the …
YOWO: You Only Walk Once to Jointly Map An Indoor Scene and Register Ceiling-mounted Cameras
F Yang, S Yamao, I Kusajima, A Moteki… - … on Circuits and …, 2024 - ieeexplore.ieee.org
Using ceiling-mounted cameras (CMCs) for indoor visual capturing opens up a wide range
of applications. However, registering CMCs to the target scene layout presents a …
of applications. However, registering CMCs to the target scene layout presents a …
Towards active vision for action localization with reactive control and predictive learning
Visual event perception tasks such as action localization have primarily focused on
supervised learning settings under a static observer, ie, the camera is static and cannot be …
supervised learning settings under a static observer, ie, the camera is static and cannot be …
Effi-MAOT: A Communication-Efficient Multi-Camera Active Object Tracking
Multi-camera Active Object Tracking (AOT) has become a promising technique to
continuously track the moving target in the intelligent surveillance system. The …
continuously track the moving target in the intelligent surveillance system. The …