Thinesh Thiyakesan Ponbagavathi — accepted AI-conference papers

2026

Frame2Freq: Spectral Adapters for Fine-Grained Video Understanding

Thinesh Thiyakesan Ponbagavathi, Constantin Seibold, Alina Roitberg

CVPR 2026

Adapting image-pretrained backbones to video typically relies on time-domain adapters tuned to a single temporal scale. Our experiments show that these modules pick up static image cues and very fast flicker changes, while overlooking medium-speed motion. Capturing dynamics across multiple time-scal

Cited by 0 Source ↗code ↗Scholar ↗

2026

Order Matters: On Parameter-Efficient Image-To-Video Probing for Recognizing Nearly Symmetric Actions

Thinesh Thiyakesan Ponbagavathi, Alina Roitberg

ICRA 2026poster

Fine-grained understanding of human actions is essential for safe and intuitive human–robot interaction. We study the challenge of recognizing nearly symmetric actions, such as picking up vs. placing down a tool or opening vs. closing a drawer. These actions are common in close human-robot collabora…

Cited by 0 Source ↗PDF ↗Scholar ↗