ZurichCV #8

Xi Wang on Egocentric Learning: Predicting the Future from a First-Person Perspective

Research in artificial intelligence continues to advance rapidly, outperforming humans in many tasks and integrating into our daily lives. However, despite their superior performance, current technologies face limitations in perceiving, processing, and understanding our visual world, particularly when it comes to understanding and interacting with people. These challenges raise the core question of my research: How do we build intelligent systems that can interact with people and offer assistance in a natural and seamless way? In this talk, I will present our work on making predictions in egocentric videos.

Nikolai Kalischek on CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generation We introduce a novel method for generating 360° panoramas from text prompts or images. Our approach leverages recent advances in 3D generation by employing multi-view diffusion models to jointly synthesize the six faces of a cubemap. Unlike previous methods that rely on processing equirectangular projections or autoregressive generation, our method treats each face as a standard perspective image, simplifying the generation process and enabling the use of existing multi-view diffusion models. We demonstrate that these models can be adapted to produce high-quality cubemaps without requiring correspondence-aware attention layers. Our model allows for fine-grained text control, generates high resolution panorama images and generalizes well beyond its training set, whilst achieving state-of-the-art results, both qualitatively and quantitatively.