The era of ultra-high-resolution imaging has arrived. Modern image sensors exceeding 200 MP resolution are common in smartphones, with over 400 MP sensors under development. However, the large number of pixels poses significant challenges for acquisition and processing, especially on edge devices.
Which pixels should be acquired, and when, for bandwidth-efficient imaging and perception?
We introduce Policy-based Foveated Imaging and Perception, an on-device, real-time, predictive, and task-aware framework that dynamically allocates sensor resolution to prioritize important regions under specific perception objectives. This paper will be presented at #SIGGRAPH2026!
[1/6]
@GordonWetzstein Who need sensors, when in that garbage quality and cpu/GPU power smartphone vendors can just put a fake camera and dream a photos based on GPS location? 😅
Ambient-robust Inverse Rendering using Active RGB-NIR Imaging
Hoon-Gyu Chung, Jinnyeong Kim, Hyunwoo Kang, Seung-Hwan Baek
(POSTECH, Republic of Korea)
https://t.co/oVOTy5b9xl
Abstract:
Inverse rendering aims to reconstruct geometry and reflectance of objects from images. Despite recent progress, existing methods often produces inaccurate reconstructions that are sensitive to ambient illumination conditions. Here we introduce an ambient-robust inverse rendering method enabled by active RGB-NIR imaging. Our key insight is to leverage near-infrared (NIR) flash illumination-imperceptible to human observers-to obtain stable point-light shading that is largely invariant to ambient illumination. By using multi-view RGB images illuminated by ambient light and NIR images acquired with active NIR flash illumination, we reconstruct accurate geometry and reflectance by exploiting the complementary benefits of RGB and NIR images via a three-stage inverse rendering method. To enable dense multi-view acquisition, we develop an active imaging system equipped with a RGB-NIR camera and a NIR flash mounted on a mobile base. Using this system, we collect the first multi-view RGB-NIR inverse rendering dataset captured under multiple ambient illumination conditions. Experiments demonstrate that our method outperforms prior approaches, achieving accurate geometry and reflectance estimation across multiple ambient lighting scenarios.
It would appear Nvidia is literally doing an Enron.
Specifically, it has reinvented the Chewco Maneuver:
Create fake "third parties" to absorb unwanted liabilities, fund with your own capital to prime the pump, then get rubes (retirement funds) to eat the risk.
OpenLogi is a native, local-first alternative to Logitech Options+ written in Rust, controlling HID++ mice without an account or telemetry.
- Interactive GUI with clickable mouse diagram and action picker
- 37 built-in actions plus recorded custom keyboard shortcuts
- DPI presets, SmartShift toggle, and per-application profile overlays
- CLI for headless device inventory and diagnostics
MeshRepair 2.8.0
with a new Remove Thin Bridges
Windows x64 and MacOS Apple Silicon binaries are free on Patreon:
https://t.co/Ryovhqbsr3
#opensource#blender#b3d
MeshRepair 2.8.0
with a new Remove Thin Bridges
Windows x64 and MacOS Apple Silicon binaries are free on Patreon:
https://t.co/Ryovhqbsr3
#opensource#blender#b3d
@MainzOnX Sources might be float or even lower. Calculation and results like color transformation matrices are in double.
This is not a realtime, and more about measurements, so no reason to not use double, even if it will be used as float at the end somewhere in image processing.