Ilya Gurvich

I'm an ambitious applied machine‑learning researcher, engineer, and a technical leader with a multidisciplinary background. In the last  17  years I've been tackling many intriguing problems in the fields of computer vision and deep learning. My background is in cyber‑security research, and I have a track record of initiating, executing, and shipping highly complex projects.

My research interests encompass machine learning - especially deep learning - applied to computer vision, audio processing, teleconferencing, and, more recently, generative multimodal AI models.

You can learn more about my professional background on LinkedIn and explore my academic publications on Google Scholar.

Highlighted Work

• Notsofar‑1 challenge: New datasets, baseline, and tasks for distant meeting transcription. Code and resources are available on GitHub.
• A real‑time active speaker detection system integrating an audio‑visual signal with a spatial querying mechanism
• Advances in online audio‑visual meeting transcription
• Toward realistic hands gesture interface: Keeping it simple for developers and machines
• Pedestrian Re‑Identification Across Two Cameras. This technical report also introduces the concept of transitive re-identification and provides its mathematical formulation (which led to this paper later).
• Learning implicit transfer for person re‑identification