Search for: All records

Visual aliasing, or doppelgangers, poses severe challenges to 3D reconstruction. We propose Doppelganger++, an enhanced pairwise image classifier that excels in visual disambiguation across diverse and challenging scenes. We seamlessly integrate Doppelganger++ into SfM, successfully disambiguating each scene. (Middle) Compared to prior work (which we refer to as DG-OG), Doppelgangers++ is more robust for everyday scenes, showing improved accuracy and robustness. We show pairs that DG-OG classifies incorrectly and ours gets correct. Our new VisymScenes dataset, featuring complex daily scenes, is particularly challenging for COLMAP and DG-OG, but our method can achieve correct and complete reconstructions. 

We introduce a method for high-quality 3D reconstruction from multi-view images. Our method uses a new point-based representation, the regularized dipole sum, which generalizes the winding number to allow for interpolation of per-point attributes in point clouds with noisy or outlier points. Using regularized dipole sums, we represent implicit geometry and radiance fields as per-point attributes of a dense point cloud, which we initialize from structure from motion. We additionally derive Barnes-Hut fast summation schemes for accelerated forward and adjoint dipole sum queries. These queries facilitate the use of ray tracing to efficiently and differentiably render images with our point-based representations, and thus update their point attributes to optimize scene geometry and appearance. We evaluate our method in inverse rendering applications against state-of-the-art alternatives, based on ray tracing of neural representations or rasterization of Gaussian point-based representations. Our method significantly improves 3D reconstruction quality and robustness at equal runtimes, while also supporting more general rendering methods such as shadow rays for direct illumination. 

Abstract This study investigates the antimicrobial effectiveness of 405 nm light emitting diodes (LEDs) against pathogenicEscherichia coliO157:H7,Listeria monocytogenes,Pseudomonas aeruginosa,SalmonellaTyphimurium, andStaphylococcus aureus, in thin liquid films (TLF) and on solid surfaces. Stainless steel (SS), high density polyethylene (HDPE), low density polyethylene (LDPE), and borosilicate glass were used as materials typically encountered in food processing, food service, and clinical environments. Anodic aluminum oxide (AAO) coupons with nanoscale topography were used, to evaluate the effect of topography on inactivation. The impact of surface roughness, hydrophobicity, and reflectivity on inactivation was assessed. A 48 h exposure to 405 nm led to reductions ranging from 1.3 (E. coli) to 5.7 (S. aureus) log CFU in TLF and 3.1 to 6.3 log CFU on different solid contact surfaces and packaging materials. All inactivation curves were nonlinear and followed Weibull kinetics, with better inactivation predictions on surfaces (0.89 ≤ R² ≤ 1.0) compared to TLF (0.76 ≤ R² ≤ 0.99). The fastest inactivation rate was observed on small nanopore AAO coupons inoculated withL. monocytogenesandS. aureus, indicating inactivation enhancing potential of these surfaces. These results demonstrate significant promise of 405 nm LEDs for antimicrobial applications in food processing and handling and the healthcare industry. 

We propose a special design enclosure device that promotes isotropic distribution of germicidal UV-C light for the effective disinfection of difficult to reach surfaces. We used experimental and computational approaches to investigate the disinfection efficacy of this device againstEscherichia coliandListeria innocua. Stainless steel, Copper metal, and a Copper polymer were used as solid substrates of varying roughness and hydrophobicity. Bacteria reductions of up to 6.9 log CFU were achieved at various locations relative to the UV-C source after 3 min of treatment (20–990 mJ/cm²cumulative fluence depending on the location). Inactivation kinetics was nonlinear and followed the Weibull model (0.77 ≤R²≤ 0.97). Optical ray tracing simulation was used to generate maps of spatial light distribution, which were then coupled with microbial inactivation kinetics to create spatial maps of inactivation. The modeling approach used accurately predicted microbial inactivation at various locations, with only small discrepancies (±8%) between predicted and experimental data. These findings demonstrate that the proposed device is suitable for disinfecting various hard to reach surfaces, with numerous possible applications in the food and healthcare industries. Additionally, the modeling approach used here can be used to aid in the design of a highly effective Ultraviolet treatment system.