Accurate camera calibration is a fundamental task for 3D perception, especially when dealing with real-world, in-the-wild environments where complex optical distortions are common. Existing methods often rely on pre-rectified images or calibration patterns, which limits their applicability and flexibility. In this work, we introduce a novel framework that addresses these challenges by jointly modeling camera intrinsic and extrinsic parameters using a generic ray camera model. Unlike previous approaches, AlignDiff shifts focus from semantic to geometric features, enabling more accurate modeling of local distortions. We propose AlignDiff, a diffusion model conditioned on geometric priors, enabling the simultaneous estimation of camera distortions and scene geometry. To enhance distortion prediction, we incorporate edge-aware attention, focusing the model on geometric features around image edges, rather than semantic content. Furthermore, to enhance generalizability to real-world captures, we incorporate a large database of ray-traced lenses containing over three thousand samples. This database characterizes the distortion inherent in a diverse variety of lens forms. Our experiments demonstrate that the proposed method significantly reduces the angular error of estimated ray bundles by ~8.2 degrees and overall calibration accuracy, outperforming existing approaches on challenging, real-world datasets.

Head-worn displays are becoming a viable visual display option for mobile augmented reality. In this paper, we highlight the challenges and some recent progress towards compact and lightweight head-worn displays tending towards the eyeglass form factor. Single and dual-element eyeglass display designs that leverage the advances in free-form optics fabrication technology will be presented. There are three new contributions in this paper: first, we present a single-element eyeglass display design and the fabricated mirror prototype; second, we present our second generation dual-element eyeglass display design with the revised opto-mechanical packaging; third, we present our initial studies on the field of view limit with an 8 mm pupil for the dual-element design.

Provides an abstract for each of the presentations and a brief professional biography of each presenter. The complete presentations were not made available for publication as part of the conference proceedings.

We are proposing an optical see-through head-worn display that is capable of mutual occlusions. Mutual occlusion is an attribute of an augmented reality display where real objects can occlude virtual objects and virtual objects can occlude real objects. For a user to achieve the perception of indifference between the real and the virtual images superimposed on the real environment, mutual occlusion is a strongly desired attribute for certain applications. This paper presents a breakthrough in display hardware from a mobility (i.e. compactness), resolution, and a switching speed based criteria. Specifically, we focus on the research that is related to virtual objects being able to occlude real objects. The core of the system is a spatial light modulator (SLM) and polarization-based optics which allow us to block or pass certain parts of a scene which is viewed through the head-worn display. An objective lens images the scene onto the SLM and the modulated image is mapped back to the original scene via an eyepiece. We are combining computer generated imagery with the modulated version of the scene to form the final image a user would see.