From the 2 of 24 linked papers with an AI index.
24 papers
Benchmarking Nighttime Traffic Sign Recognition with Illumination-Adaptive Detection and Semantic Attribute Reasoning
Aditya Mishra, Akshay Agarwal, Haroon Lone
The paper presents INTSD, a large-scale nighttime traffic sign dataset collected across India, and introduces LENS-Net, an illumination‑adaptive detector with a vision‑language cla…
Simple Features and Honest Calibration for Ambivalence and Hesitancy Recognition in Video
Vikas Kumar, Aditya Mishra, Haroon R. Lone
The paper presents a system for detecting ambivalence and hesitancy in short interview videos by combining text, audio, and visual cues with a simple fusion method and an AP-weight…
Context-Dependent Autonomic Responses in Social Anxiety During Cognitive-Emotional Stress
Arya Adyasha, Vikas Kumar, Anushka Sanjay Shelke +1
Social anxiety disorder (SAD) is associated with heightened physiological arousal during socially evaluative situations, yet it remains unclear whether similar autonomic responses…
Beyond Benchmarks: Continuous Edge Inference for Fine-Grained Roadside Perception
Aditya Mishra, Haroon Lone
Continuous AI inference on resource-constrained edge hardware introduces deployment effects that are largely invisible to conventional benchmark evaluation, including temporal inst…
Q-LocalAdam: Memory-Efficient Client-Side Adaptive Optimization for Edge Federated Learning
Vedant Waykole, Haroon R. Lone
Federated learning on edge devices must cope with non-IID client data and tight memory budgets. Adaptive optimizers like Adam stabilize training under data heterogeneity but requir…
FedPrism: Adaptive Personalized Federated Learning under Non-IID Data
Prakash Kumbhakar, Shrey Srivastava, Haroon R Lone
Federated Learning (FL) suffers significant performance degradation in real-world deployments characterized by moderate to extreme statistical heterogeneity (non-IID client data).…