3D Gaussian SplattingRead 2026-11-06
Jinxin Ai, Matthias Nießner · arXiv (Cornell University) · 2026
GraftRay inserts mesh objects into indoor scenes reconstructed with 2D or 3D Gaussian Splatting, producing lighting and shadows that stay consistent across views. It recovers scene materials and normals with diffusion-based priors, models spatially extended emitters such as windows and lamps, and jointly path-traces the Gaussians and inserted meshes. On the FIPT benchmark, it reports state-of-the-art rendering fidelity and shadow accuracy over ComGS and EAG-PT.
phase change materialRead 2026-11-06
Haoran Zhang · Journal of Civil and Transportation Engineering · 2026
This review surveys phase change materials (PCMs) for building energy efficiency, covering organic, inorganic, and eutectic PCMs, along with macro-, micro-, and shape-stabilized encapsulation methods. It reports that PCMs can cut annual heat gain by 21–23% in tropical climates and reduce Mediterranean cooling loads by 7.5–9.5% and heating loads by 55–61.6%. The authors also identify remaining challenges in long-term stability, thermal conductivity, cost, and simulation-experiment agreement.
neural radiance fieldsRead 2026-11-06
Nur Aini Lestari, Harintaka Harintaka, Ahmad Rasis Sholikin · 2026
This paper presents a semi-automatic workflow for building LoD2 3D city models, in which users digitize roof outlines from orthophotos and label roof types (flat, gable, hip, shed, pyramid), and rule-based code then reconstructs roof geometry using height and orientation derived from point cloud elevation layers. Applied to 371 buildings, the method outputs CityJSON models convertible to CityGML. On 30 buildings, roof-plane completeness averaged 100% and correctness and quality averaged 98.33%, with a geometric RMSE of 0.90 m.
Open access science paperRead 2026-11-05Cited source
G. F. Moore, Suzanne Audrey, Mary Elizabeth Barker, Lyndal M Bond et al. · BMJ · 2015
This MRC guidance sets out a framework for designing, conducting, and reporting process evaluations, which examine how complex interventions are implemented and how they work in practice. The authors argue that process evaluation is an essential component of testing complex interventions. The source text provides only the abstract, so details of the framework itself are not described here.
Open access science paperRead 2026-11-05Cited source
David Nicol, Debra Macfarlane · Studies in Higher Education · 2006
This paper reframes research on formative assessment and feedback around self-regulated learning, arguing that students already generate their own feedback and that higher education should build on this. From this model it derives seven principles of good feedback practice, with supporting research and easy-to-implement strategies. The authors contend that treating students as proactive rather than reactive feedback users has significant implications for how teachers design assessment.
agrivoltaicsRead 2026-11-05
White, Robin R., Hall, Ralph P., Ignosh, John, Fike, John H. et al. · VTechWorks (Virginia Tech) · 2025
This document reports on a 2024 summit in Virginia focused on agrivoltaics, the dual use of farmland for agriculture and solar energy production. The summit aimed to identify collaboration strategies between industry and state agencies, examine opportunities and challenges from varied perspectives, and form a network to advance agrivoltaics for Virginia stakeholders. The abstract describes the summit's goals but does not report specific findings or outcomes.
LoRa LoRaWanRead 2026-11-05
Rushikesh S. Desale · International Journal for Research in Applied Science and Engineering Technology · 2025
This survey reviews Internet of Things approaches to smart streetlight monitoring and automated fault detection, examining architectures that combine embedded sensors, microcontrollers, and cloud platforms. The reviewed systems use light, motion, and power sensors to detect malfunctions and adjust illumination to conditions and traffic. The authors report that such networks can reduce energy use and speed maintenance response, while noting open challenges in network reliability, scalability, and cost.
Open access science paperRead 2026-11-05
Agneta Malmgren Fänge, Vigdis Abrahamsen Grøndahl, Helene Åvik Persson, Heidi Kristine Grønlien et al. · Advances in Medical Education and Practice · 2026
This study evaluated the design of the NorDigHE program, a self-paced, multimodal digital learning course for Hospital-at-Home care, using usability scores and participant surveys from 125 students and healthcare professionals in Denmark, Norway and Sweden. Usability exceeded the established cut-off, and podcasts, case videos, simulations and quizzes were generally rated as relevant and supportive of learning. Some learners found simulations insufficiently challenging, pointing to a need to tailor difficulty.
3d resin printingRead 2026-11-04
Nabilah Hassim, Siti Nadia Rahimi, Faezah Rokhani, Muhammad Syafiq Alauddin et al. · Progress in Additive Manufacturing · 2026
This laboratory study compared a flexible 3D-printed dental resin, fabricated with either digital light processing (DLP) or masked stereolithography (MSLA), against conventional thermoformed ethylene vinyl acetate at 4 mm thickness. Both printed groups showed significantly higher tensile strength, elastic modulus, flexural properties, and fracture toughness than thermoformed EVA, with the MSLA group performing highest. The findings provide baseline mechanical data for this comparison.
machine learningRead 2026-11-04
Jiling Zhang, Xinjun Zhang, Yandan Wu · Frontiers in Medicine · 2026
This study tested whether routine complete blood count values could estimate the hemoglobin overestimation caused by lipemia, using plasma-replacement hemoglobin as the reference, in 283 clinical specimens. A random forest model using routine CBC parameters predicted the relative bias well (out-of-fold R² of 0.848) and produced corrected hemoglobin values closely matching the reference (R² of 0.975). The authors note that accuracy was lower in the highest-bias cases, and that the model is not ready to replace established analytical correction methods without independent prospective validation.
residential solar thermal heating systemsRead 2026-11-04
Elahe Moazzen, Mohammad Zare-Anbouhi, Mohammad Mohsen Hayati, Francesco Guarino et al. · Scientific Reports · 2026
This study links retrofit optimization with district energy scheduling for a 17-story residential tower. An EnergyPlus model coupled with NSGA-II selected insulation, glazing, shading, and setpoint changes that cut annual energy use by 14.85% and lowered average predicted percentage dissatisfied from 29.51% to 18.69%. Scaled to 816 units in an energy hub model, storage with demand response reduced operating cost by 19.90%.
neural radiance fieldsRead 2026-11-04
Sarah Zhang, Zhiyang Dou, Minghao Guo, Benjamin Tod Jones et al. · arXiv (Cornell University) · 2026
RADIANT is a feed-forward neural renderer that synthesizes global illumination by representing each object with a fixed budget of 32 surface-patch tokens, so attention cost depends on object count rather than mesh resolution. It uses an invariant form-factor attention to model light exchange between objects and a cross-attention view decoder to render images. On a 1.03M-scene benchmark, it reaches 32.2 dB PSNR at 600k training steps, 3.3 dB above RenderFormer.
phase change materialRead 2026-11-04
张子祺 · uO Research (University of Ottawa) · 2026
This thesis develops phase-change-material-enhanced hemp-lime composites (PCM-HLC) by impregnating hemp shiv with bio-derived PCMs and designing binders with supplementary cementitious materials to support up to 70% PCM substitution. The 70% composite showed up to 80% higher specific heat capacity during phase transition, but at the cost of lower compressive strength and a 66% drop in moisture buffering value. Numerical modelling of a multilayer panel for a cold climate found an optimised 10 cm design that eliminated the modelled summer cooling demand and cut winter heating energy by 27% versus a monolithic HLC baseline.
3D Gaussian SplattingRead 2026-11-03
Seongjun Choi, Sangyoun Lee · arXiv · 2026
RelateGS is a language-aware 3D Gaussian Splatting method for 3D visual grounding that aims to make semantic and relational evidence more reliable. It suppresses unreliable observations during semantic feature uplifting, consolidates multi-view object observations into a persistent 3D graph, and uses that graph to re-query existing frames for under-observed relations. On Sr3D+ and Nr3D, it improves the previous best Acc@0.25 by 7.9 and 1.7 percentage points, respectively.
LoRa LoRaWanRead 2026-11-03
Vineet Wankhede, Sanwedana Lokhande, Sejal Vaidya, Aditya Khandalkar · International Journal of Scientific Research in Engineering and Management · 2026
SkySense is a proposed quadcopter-based weather and air-quality monitoring system that streams sensor readings (DHT22 temperature and humidity, ENS160 air purity) to a ground station over a point-to-point LoRa link, where an ESP32 displays data and forwards it to Google Cloud. The paper's contribution is the architecture, packet design, and analytical estimates of airtime, link margin, and hover endurance, along with a verification protocol. Its features remain classified as proposed, pending field trials.
3D Gaussian SplattingRead 2026-11-03
Toivo Ylinampa, Urmas Kõljalg · Biodiversity Data Journal · 2026
This paper presents a practical workflow that combines photogrammetry (Structure-from-Motion and Multi-View Stereo) with 3D Gaussian Splatting to digitise plants at scales from single flowers to greenhouse environments, using consumer-grade cameras. Tested on a 12-timepoint fern growth series and a greenhouse digital twin, it reports photorealistic rendering quality (PSNR 21.99–26.35 dB) and shows that leaf movement from wind markedly reduces reconstruction fidelity.
freecadRead 2026-11-03
Mariona Royes Moltó · Repositori ObertUDL (University of Lleida) · 2026
This master's thesis develops a low-cost, adaptive generative design system for transtibial prosthetic sockets, aimed at Gambian users who lack access to traditional plaster casting. The workflow combines 3D scanning of the residual limb, parametric biomechanical adjustments in Rhinoceros and Grasshopper, and FFF printing with ASA and TPU. The reported result is a design validated against ISO 10328 P6 loading, produced at about 31.26 € per socket.
esp32-s3Read 2026-11-03
Jessica Wang · 2026
This paper measures how much one VL53L1X time-of-flight sensor is disturbed by an identical neighbor 20 mm away, as a function of the angle between them, using one sensor pair against a gray target. The extra ambient light falls to about 10 percent of its peak at roughly 16 to 18 degrees total angle for a full array, and the neighbor's ranging was barely shifted, though near the acceptance threshold it could change the share of accepted readings by up to about 20 percentage points.
residential solar thermal heating systemsRead 2026-11-02
Umidjon Ibragimov, Temirova Lobar Zokirovna, U. Mirzayorova Sevara · Innovation Science and Technology · 2026
This paper reviews how electric underfloor heating combined with autonomous power sources, such as photovoltaic panels, thermal storage, and phase-change materials, could make residential heating more energy-efficient. Drawing on international research, it argues that low-temperature underfloor systems can reduce energy use and heat loss compared with radiator or forced-air systems while improving thermal comfort. The abstract presents these advantages as promising rather than quantified, so specific savings figures are not stated here.
3D Gaussian SplattingRead 2026-11-02
Quan Tran Hong, Long Nguyen-Chi, Binh Thanh Nguyen · arXiv (Cornell University) · 2026
This paper proposes replacing heuristic density control in 3D Gaussian Splatting with a variational score derived from the multi-view reconstruction loss, computed by differentiating the loss with respect to auxiliary mass coordinates on each frozen Gaussian. High-score Gaussians are chosen as birth sites and low-score, low-opacity ones for removal. Integrated into 3DGS-MCMC, the authors report state-of-the-art reconstruction quality across standard benchmarks, with the largest gains under tight Gaussian budgets.
3D Gaussian SplattingRead 2026-11-02
Wenqing Wang, Huimin Zeng, Yun Fu · arXiv (Cornell University) · 2026
RacerGS compresses 3D Gaussian Splatting scenes using a hash projection model, a FiLM-modulated autoregressive GRU context model, and residual entropy coding of anchor attributes. The authors report over 150x compression relative to vanilla 3DGS and 28x relative to Scaffold-GS, with comparable rendering quality across five datasets.
Open access science paperRead 2026-11-01
Swaechchha Dahal, Sambeet Mishra, Gunne John Hegglid, Bhupendra Bimal Chhetri et al. · Scientific Reports · 2026
The paper proposes WINDS, a power system dispatch framework that pairs neural learning for weather-driven uncertainty with symbolic reasoning that enforces security constraints. In tests on a controlled experiment and measured wind and load data from Denmark's DK1 zone, it reports operational cost reductions of 11.4% to 26.7% versus climatology-based dispatch and over 80% fewer constraint violations than forecast-only dispatch. The results are presented as a proof of concept.
esp32-s3Read 2026-11-01
Alexandre Nodari, Marcelo Luis Walter, Leonardo Nunes, Marcelo E. Pellenz et al. · Lecture Notes in Business Information Processing · 2026
This experience report describes a university-industry partnership that developed an AIoT product using a Multi-Cycle Design Science Research framework. The framework runs fixed-length cycles in which each cycle's main deliverable is a peer-reviewed publication that also tackles a product development challenge. Over four cycles, the authors report progressively reducing project risk and moving toward a commercially viable product, while producing four academic articles.
phase change liquidRead 2026-11-01
Natalia Kołton, Piotr Jeleń, Jakub Marchewka, Marek Nocuń et al. · Journal of Thermal Analysis and Calorimetry · 2026
This study develops bio-based thermochromic phase change materials from a three-component mixture of bromocresol purple dye, stearic acid, and behenyl alcohol, where the fatty acid and alcohol both serve as phase change materials. The mixtures show high melting enthalpy (about 150–200 J/g) between 50 and 70 °C, thermal stability up to roughly 190 °C, and a clear color change during phase transitions, suggesting use as both energy storage media and colorimetric indicators.
3D Gaussian SplattingRead 2026-11-01
Xiaobin Deng, Changyu Diao, Min Li, 许端清 · arXiv (Cornell University) · 2026
BoundedGS is a memory-efficient training method for high-resolution 3D Gaussian Splatting. It processes each view in bounded local regions, separating rasterization and projection backward passes so that cached data can be released after each region contributes its gradients. On five native 5K outdoor scenes, it reduced the mean rendering working set from 9.97 to 1.76 GiB (82.3%) while keeping similar reconstruction quality and cutting training time from 36.7 to 33.9 minutes.
Open access science paperRead 2026-10-31
Francesca Tonolo, Vivek Vijay, Ernesto Quintas Sánchez, Adrian Batista-Planas et al. · Astronomy and Astrophysics · 2026
This paper provides the first state-to-state collisional rate coefficients for E- and Z-ethanimine, a prebiotically relevant molecule, in collisions with helium and para-H2 over 5–100 K. The rates were computed from quantum scattering calculations for He and extended to para-H2 using scaling factors derived from comparisons with mixed quantum-classical calculations. The datasets are intended to support non-LTE modeling of ethanimine in cold interstellar clouds and future searches beyond Sgr B2(N).
3D Gaussian SplattingRead 2026-10-31
Xiaobin Deng, Changyu Diao, Min Li, 许端清 · arXiv (Cornell University) · 2026
ParetoGS is a method for scalable 3D Gaussian Splatting that serves many Gaussian-count budgets from one representation. It builds a small set of adapted budget anchors, stores their parameter changes in a compact residual atlas, and interpolates between anchors to decode intermediate budgets without separate optimization for each. Across 13 scenes, it reports better average PSNR, SSIM, and LPIPS than compared prefix methods at 10% of each method's full model, and it cuts mean additional state from 390.0 to 85.7 MB with a 0.01 dB average PSNR change.
phase change materialRead 2026-10-31
Hussein Alawai Ibrahim Al-Saaidi, Maha A. Hussein, Magdalena Barnetche, Cristina Prieto · Journal of Energy Storage · 2026
This study uses a three-dimensional CFD model in ANSYS Fluent to examine how non-uniform, axially sectorised electrical heating affects charging of a metallic phase change material storage tank used in direct steam generation for concentrated solar power. The model, validated against literature data, shows that gradient heating improves thermal homogeneity, reduces local overheating, and improves charging efficiency while lowering the total power input compared with uniform heating.
freecadRead 2026-10-31
최현덕 · 2026
This paper describes a lightweight indoor cellular signal simulator built from high school physics equations and ray tracing, applied to a university dormitory floor plan modeled in FreeCAD and Blender. Signal strength was measured at nine indoor and two outdoor locations and compared against the simulator and an electromagnetic simulator. The program achieved a mean absolute error of 6.57 dB, slightly better than the 7.78 dB of the reference simulator, and identified the two weakest measured locations among its three lowest estimates.
esp32-s3Read 2026-10-31
Shreyash Panchal · INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · 2026
This review examines a low-cost wireless sensor network built on LoRa radio, using ESP32 or Arduino microcontrollers and sensors such as PIR, magnetic contact, and vibration detectors to monitor remote perimeters and report intrusions to a central station. The authors survey the system architecture, communication principles, and practical challenges, including interference, packet loss, security, scalability, and power management. They conclude that LoRa-based networks are a promising option for non-weaponized perimeter monitoring, provided reliability, cybersecurity, and regulatory requirements are addressed.