Tech Papers 2026: This paper applies Judea Pearl’s causal theory to AI-assisted editing, proposing a framework that treats the cut as a deliberate causal intervention to help AI understand narrative intent and support creative decision-making.
Abstract
As AI video tools advance from simple automation to agentic control, the industry faces a critical "Narrative Gap": models can map surface-level patterns, but they cannot reason about the structural intent behind an edit sequence.
You are not signed in
Only registered users can read the rest of this article.
Evolving the codec: Modern video delivery with AV2
Tech Papers 2026: This paper examines AV2, a next-generation video codec designed to improve compression efficiency while supporting a wider range of use cases.
Osmia VP: A low-cost, field-deployable virtual production system for in-camera VFX
Tech Papers 2026: This paper examines a low-cost virtual production system designed to make in-camera VFX more accessible to mid- and low-budget productions using commodity hardware.
Implementing a single-player multiview solution
Tech Papers 2026: This paper examines four approaches to multiview delivery, comparing their flexibility, efficiency, and scalability before exploring the benefits of single-decoder operation.
Agentic AI for smart TV QA: A production field study across multiple platforms and languages
Tech Papers 2026: This paper examines how agentic AI can expand Smart TV test coverage by overcoming the maintenance limits of conventional automation in a constantly changing environment.
Video-only camera pose estimation for sports broadcast graphics using venue-aware deep learning
Tech Papers 2026: This paper presents a video-only camera pose estimation framework that eliminates the need for dedicated tracking hardware.