Jordan Baker | Searcharxiv

arXiv2026

From Fog Chamber to Aircraft Window: Pixel-Registered Imaging and Synthetic Fine-Tuning Enable Cross-Domain Defogging

A deep defogging pipeline pretrained on controlled laboratory fog and fine-tuned with domain-randomized synthetic fog applied to clear outdoor scenes generalizes across a graded sequence of out-of-distribution settings with no target-domain training, from chamber-free free-flowing fog to iPhone video recorded through an aircraft cabin window in flight, an entirely unseen sensor, scene, and optical path. This directly addresses an open transfer limitation reported for real-world binocular defogging. Two design choices support the transfer. First, a single-camera fog imager photographs a flat-panel display through an artificial-fog enclosure with a fixed 114~mm scattering path, producing 5{,}495 pixel-aligned foggy/clear pairs. Exact registration permits a paired Laplacian ratio that predicts per-image restoration quality far better than single-image proxies (Spearman $\rho = 0.632$ versus $0.399$) and supports pixel-exact $L_1$ reconstruction training that avoids adversarial hallucination. Second, the fog-chamber checkpoint is fine-tuned on Mapillary Vistas crops overlaid with on-the-fly randomized synthetic fog spanning a broad range of strengths, spatial variations, airlights, and noise conditions. On a 552-image held-out split, a uniform comparison of 30 restoration backbones places NAFNet at the top (24.33~dB~/~0.7912~SSIM), with a compact alternative within 1.29~dB at 3\% of the parameter count, and a ResNet-50 classifier confirms that the restoration preserves semantic content rather than only pixel-level structure. On unpaired aircraft-window video, NIQE decreases from a mean of 6.22 to 4.97 after fine-tuning, with temporally stable output across full-motion sequences. The same backbone, under paired supervision, also reaches 20.71~dB~/~0.683~SSIM on a non-overlapping O-HAZE/NH-HAZE split (a transferability check rather than a competitive ranking).

cs.CV↗

arXiv2022

Dust Accretion onto Exoplanets

Accretion of interplanetary dust onto gas giant exoplanets is considered. Poynting-Robertson drag causes dust particles from distant reservoirs to slowly inspiral toward the star. Orbital simulations for the three-body system of the star, planet, and dust particle show that a significant fraction of the dust may accrete onto massive planets in close orbits. The deceleration of the supersonic dust in the planet's atmosphere is modeled, including ablation by thermal evaporation and sputtering. The fraction of the accreted dust mass deposited as gas-phase atoms is found to be large for close-in orbits and massive planets. If mass outflow and vertical mixing are sufficiently weak, the accreted dust produces a constant mixing ratio of atoms and remnant dust grains below the stopping layer. When vertical mixing is included along with settling, the solutions interpolate between the mixing ratio due to the meteoric source above the homopause, and that of the well-mixed deeper atmosphere below the homopause. The line opacity from atoms and continuum opacity from remnant dust may be observable in transmission spectra for sufficiently large dust accretion rates, a grain size distribution tilted toward the blowout size, and sufficiently weak vertical mixing. If mixing is strong, the meteoric source may still act to augment heavy elements mixed up from the deep atmosphere as well as provide nucleation sites for the formation of larger particles. The possible role of the Lorentz drag force in limiting the flow speeds and mixing coefficient for pressures $\la 1\, \rm mbar$ is discussed.

astro-ph.EP↗