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.

Recent extended reality headsets and field robots have adopted covers to protect the front-facing cameras from environmental hazards and falls. The surface irregularities on the cover can lead to optical aberrations like blurring and non-parametric distortions. Novel view synthesis methods like NeRF and 3D Gaussian Splatting are ill-equipped to synthesize from sequences with optical aberrations. To address this challenge, we introduce SynthCover to enable novel view synthesis through protective covers for downstream extended reality applications. SynthCover employs a Refractive Field that estimates the cover's geometry, enabling precise analytical calculation of refracted rays. Experiments on synthetic and real-world scenes demonstrate our method's ability to accurately model scenes viewed through protective covers, achieving a significant improvement in rendering quality compared to prior methods. We also show that the model can adjust well to various cover geometries with synthetic sequences captured with covers of different surface curvatures. To motivate further studies on this problem, we provide the benchmarked dataset containing real and synthetic walkable scenes captured with protective cover optical aberrations.

With the aging of the population, caregiving for the elderly has become an urgent social topic. While the rapid development of data-driven technologies provides tremendous promises to deal with this issue, the gap between data-driven and practical healthcare challenges their effectiveness. With the aim of providing strong data basis for the large-scale data-driven healthcare, this paper proposes a human-space-service based method, named CareMap, to model and quantity the caregiving process for the elderly aging at home. We build the intelligent individual profile to model and compute seniors’ health conditions on the one hand and care network profile to model and quantify the care resources around the elderly on the other hand. We design a CareMap based prototype to illustrate the possible application and discuss its potentials and limitations.

In this paper, we propose a method for urban building damage detection from multitemporal high resolution images using spectral and spatial information combined. Given the spectral similarity between damaged and undamaged areas in the images, two spatial features are used in the damage detection, i.e. invariant moments and LISA (local indicator of spatial association) index. These two spatial features were computed for each image object, which is produced by image segmentation. The One-Class Support Vector Machine (OCSVM), a recently developed one-class classifier was used to classify the multitemporal data to obtain building damage information. The uses of spectral data alone and plus obtained spatial features for building damage detection were separately evaluated using bitemporal Quickbird images acquired in Dujiangyan area of China, which was heavily hit by the Wenchuan earthquake. The results show that the combined use of spectral and spatial features significantly improved the damage detection accuracy, compared to that of using spectral information alone.