arXiv subjects

Asanka Perera

Publications and source records attributed to Asanka Perera.

6 recordsLinked to original sources

arXiv2026

HiLRP: Toward One Trustworthy Explanation for Vision Transformer: Conservation-Valid Attribution via Attention Primitives

Vision Transformer (ViT) design has become increasingly diverse, with backbones combining convolutional stems, windowed, linear, or multi-axis attention, patch merging, and spatial reduction in various configurations. This diversity poses challenges for existing attribution methods, whose assumptions often do not hold across ViT variants: Grad-CAM requires a terminal spatial feature map, attention rollout assumes global softmax attention, and layer-wise relevance propagation (LRP) requires module-specific rules. To the best of our knowledge, no existing method provides a unified attribution framework across this architectural space. We show that this architectural diversity can be captured by a simpler underlying structure. The attention and resolution-reduction operators in current ViTs can be decomposed into four operation types: linear maps, bilinear mixing, normalization or gating, and reindexing. Each operation admits a relevance rule that satisfies conservation. Based on these rules, HiLRP supports new backbones by construction rather than by architecture-specific derivation, and its attribution maps decompose the prediction rather than relying on heuristic assumptions. We prove conservation and conditional equivariance and verify both to machine precision. Across 14 attribution methods and 10 architectures, we find that no prior method remains reliable across ViT families, while Faithfulness Correlation becomes uninformative for backbones robust to spatial masking. HiLRP alone preserves conservation across windowed, spatial-reduction, multi-axis, and linear-attention models, where naive extensions can produce zero or inflated relevance. It also localizes attribution failures in class activation mapping, achieving 0.97 Pointing compared with 0.55 for competing methods on EfficientViT.

cs.CV↗

arXiv2026

Does Explainability Transfer? A Controlled Benchmark of Attribution Methods on Vision Transformers and CNNs

Most evidence on the effectiveness of explainable artificial intelligence (XAI) attribution methods has been established on convolutional neural networks (CNNs), with limited investigation into whether these conclusions generalize to the diverse Vision Transformer (ViT) architectures that now dominate computer vision. This paper presents a controlled benchmark that evaluates attribution quality across five dimensions: faithfulness, localization, robustness, complexity, and computational cost. A standardized framework assesses 13 attribution methods from four algorithmic families on eight representative backbones spanning CNNs, isotropic ViTs, hierarchical transformers, hybrid architectures, and linear-attention transformers. The results show that attribution performance is strongly architecture-dependent and that rankings established on CNNs do not reliably transfer to transformer-based models. CAM-based methods achieve the highest scores under the conventional bounding-box localization metric on CNNs and most ViTs but perform poorly on linear-attention architectures. Pixel-level dense-mask evaluation further reveals that these gains largely reflect metric saturation rather than accurate localization. CAM-based methods also exhibit limited robustness on global-attention transformers, whereas attention rollout provides consistently stable explanations with poor localization. Furthermore, faithfulness correlation offers limited discrimination between attribution methods, highlighting the limitations of single-metric evaluation. These findings challenge prevailing conclusions on attribution performance and demonstrate the need for architecture-aware, multi-dimensional evaluation. The open-source code for the evaluation framework and benchmark results is available at https://github.com/Nishan-Charlie/VIT_XAI_Bench.

cs.CV↗

arXiv2026

Machine Learning Challenges in Intelligent Unmanned Aerial Vehicle Operations in Developing Economie

Unmanned aerial vehicle (UAV) environments present significant challenges for machine learning (ML) due to limited platform resources, heterogeneous sensor data, dynamic mission conditions, and safety-critical requirements. This paper examines these constraints across the core functional areas of UAV intelligence, including navigation, perception, communication-aware operation, and resilience specifically in the context of developing economies. In such settings, these challenges are often amplified by constraints such as cost sensitivity, limited infrastructure, intermittent connectivity, regulatory uncertainty, and harsh or variable operating environments. The discussion highlights the gap between ML performance in controlled experimental backgrounds and dependable deployment in real-world UAV missions within developing economies context.

eess.SY↗

arXiv2026

Benchmarking CNN and Transformer-Based Object Detectors for UAV Solar Panel Inspection

Timely and accurate detection of defects and contaminants in solar panels is critical for maintaining the efficiency and reliability of photovoltaic (PV) systems. While recent studies have applied deep learning to PV inspection, fair benchmarking across detector architectures and unbiased handling of class imbalance remain limited. This work presents a comprehensive benchmark of convolutional and transformer-based object detectors on UAV-captured RGB imagery of solar panels. It introduces a class-targeted augmentation strategy applied exclusively to the training split to mitigate imbalance without compromising evaluation integrity. Faster R-CNN with ResNet50 and MobileNetV3 backbones, RetinaNet with ResNet50, YOLOv5, YOLOv8, and Swin Transformer backbones integrated with Faster R-CNN (Tiny, Small, and Base variants) are evaluated. Performance is assessed using mean Average Precision (mAP) across multiple IoU thresholds, precision, recall, F1 score, and inference throughput to enable accuracy-throughput tradeoff analysis relevant to UAV deployment. Experimental results show that Faster R-CNN with a ResNet50 backbone achieves the highest localization accuracy, with mAP@0.5 of 0.893 and mAP@0.5:0.95 of 0.759, whereas the MobileNetV3 variant provides the best overall reliability balance, achieving recall of 0.745, F1-score of 0.809, and accuracy of 0.679 on the test set. The dataset and code will be released upon acceptance of the paper.

cs.CV↗

arXiv2023

Radio Source Localization using Sparse Signal Measurements from Uncrewed Ground Vehicles

Radio source localization can benefit many fields, including wireless communications, radar, radio astronomy, wireless sensor networks, positioning systems, and surveillance systems. However, accurately estimating the position of a radio transmitter using a remote sensor is not an easy task, as many factors contribute to the highly dynamic behavior of radio signals. In this study, we investigate techniques to use a mobile robot to explore an outdoor area and localize the radio source using sparse Received Signal Strength Indicator (RSSI) measurements. We propose a novel radio source localization method with fast turnaround times and reduced complexity compared to the state-of-the-art. Our technique uses RSSI measurements collected while the robot completed a sparse trajectory using a coverage path planning map. The mean RSSI within each grid cell was used to find the most likely cell containing the source. Three techniques were analyzed with the data from eight field tests using a mobile robot. The proposed method can localize a gas source in a basketball field with a 1.2 m accuracy and within three minutes of convergence time, whereas the state-of-the-art active sensing technique took more than 30 minutes to reach a source estimation accuracy below 1 m.

cs.RO↗

arXiv2023

Coverage Path Planning with Budget Constraints for Multiple Unmanned Ground Vehicles

This paper proposes a state-machine model for a multi-modal, multi-robot environmental sensing algorithm. This multi-modal algorithm integrates two different exploration algorithms: (1) coverage path planning using variable formations and (2) collaborative active sensing using multi-robot swarms. The state machine provides the logic for when to switch between these different sensing algorithms. We evaluate the performance of the proposed approach on a gas source localisation and mapping task. We use hardware-in-the-loop experiments and real-time experiments with a radio source simulating a real gas field. We compare the proposed approach with a single-mode, state-of-the-art collaborative active sensing approach. Our results indicate that our multi-modal switching approach can converge more rapidly than single-mode active sensing.

cs.MA↗