A survey on approximate edge AI for energy efficient autonomous driving services

D Katare, D Perino, J Nurmi, M Warnier… - … Surveys & Tutorials, 2023 - ieeexplore.ieee.org
Autonomous driving services depends on active sensing from modules such as camera,
LiDAR, radar, and communication units. Traditionally, these modules process the sensed …

[HTML][HTML] Addressing the Gaps of IoU Loss in 3D Object Detection with IIoU

N Ravi, M El-Sharkawy - Future Internet, 2023 - mdpi.com
Three-dimensional object detection involves estimating the dimensions, orientations, and
locations of 3D bounding boxes. Intersection of Union (IoU) loss measures the overlap …

Analyzing and mitigating bias for vulnerable classes: Towards balanced representation in dataset

D Katare, DS Noguero, S Park, N Kourtellis… - arxiv preprint arxiv …, 2024 - arxiv.org
The accuracy and fairness of perception systems in autonomous driving are essential,
especially for vulnerable road users such as cyclists, pedestrians, and motorcyclists who …

Approximating vision transformers for edge: variational inference and mixed-precision for multi-modal data

D Katare, S Leroux, M Janssen, AY Ding - Computing, 2025 - Springer
Vision transformer (ViTs) models have shown higher accuracy, robustness and large volume
data processing ability, creating new baselines and references for perception tasks …

Approximation Strategies for Vision Models on Edge Devices: An Accuracy-Efficiency Trade-off

D Katare, S Shakibhamedan, N Amirafshar… - Authorea …, 2024 - techrxiv.org
Autonomous applications having AI models and algorithms as backbone require high-
performance computational and memory resources for efficient deployment and data …