3D computer graphics
PulseAugur coverage of 3D computer graphics — every cluster mentioning 3D computer graphics across labs, papers, and developer communities, ranked by signal.
6 day(s) with sentiment data
-
New BEV3SR Framework Enhances 3D Spatial Reasoning in Vision-Language Models
Researchers have introduced BEV3SR, a novel framework designed to enhance 3D spatial reasoning capabilities in Vision-Language Models (VLMs) when processing video data. This training-free and backbone-agnostic system tr…
-
LucidLyra framework enables queryable 3D scene reconstruction from video
Researchers have developed LucidLyra, a novel framework for 3D scene reconstruction that allows for semantic and instance querying directly from video latents. This approach distills knowledge from video diffusion model…
-
New benchmark SpatialScout reveals VLM limitations in 3D mapping
Researchers have introduced SpatialScout, a new benchmark designed to evaluate the 3D mapping capabilities of vision-language model (VLM) agents. The benchmark consists of 80 indoor scenes with over 3,000 objects, requi…
-
SUPRA-Tex generates high-fidelity 3D materials from single images
Researchers have developed SUPRA-Tex, a novel method for generating high-fidelity 3D materials from single images. This approach decouples material estimation, surface alignment, and 3D completion to preserve intricate …
-
Tachuh launches first 3D scanner, aiming to simplify reality digitization
Tencent Technology (Tachuh) has announced its first 3D scanner, the S1, set to launch in November. This move marks the company's expansion into the "reality digitization" sector, broadening its product portfolio beyond …
-
New method assesses point cloud quality without reference
Researchers have developed a new method for assessing the quality of point clouds without needing a reference. This approach calibrates reconstruction discrepancies by estimating and compensating for the influence of th…
-
GenIA framework improves 3D object reconstruction by aligning generative models with visual data
Researchers have introduced GenIA, a novel framework designed to improve 3D object reconstruction from limited visual input. GenIA aligns generative 3D foundation models with observed geometry and appearance data withou…
-
Robotic agents' co-evolved communication struggles to transfer from 2D to 3D simulations
Researchers have investigated the transferability of communication mechanisms developed by co-evolved robotic agents from 2D simulations to 3D environments. Studies indicate that while some communication functions persi…
-
CALIPER framework uses RGB-D data for industrial part recognition
Researchers have developed CALIPER, a novel framework for recognizing visually similar industrial parts without requiring large, class-specific datasets or CAD models. This model-free approach utilizes RGB-D data to com…
-
New 3D Generation Methods Enable Part-Level Control and Selective Output
Researchers have developed Kaininja, a novel method that extends existing native 3D generators to produce part-level outputs. This is achieved through a dual-volume representation that addresses the challenge of represe…
-
ECCV 2026: AI research spans 3D reconstruction, urban, and thermal imaging
Researchers presented work at the European Conference on Computer Vision (ECCV) covering topics such as 3D reconstruction, urban environments, and thermal imaging. The conference is scheduled to reconvene in two years.
-
Bilibili's AI competition showcases creativity, with over 80% solo teams · 1 source tracked
Bilibili's first AI Creation Open Competition concluded, awarding the top prize of 1 million yuan to "Project N.E.K.O." for its AI-generated catgirl character. The competition saw over 30,000 submissions, with more than…
-
3D bone geometry reconstructed from two X-rays using statistical shape model
Researchers have developed a novel method to reconstruct 3D bone geometry from just two X-ray images, bypassing the need for CT scans or large neural network datasets. The approach utilizes a statistical shape model der…
-
Robotic platform uses active sensing for detailed plant stress analysis
Researchers have developed an autonomous robotic platform that integrates active physiological sensing, specifically chlorophyll fluorescence, with 3D plant reconstruction and geometric analysis. This system uses motion…
-
Stream3Dv2 framework enhances zero-shot 3D scene understanding with geometric-semantic fusion
Researchers have developed Stream3Dv2, a novel framework for robust, zero-shot 3D scene understanding using streaming RGB-D inputs. This training-free approach addresses limitations in handling sequential data and noise…
-
Maker Tool market poised for growth amid cash flow and tech challenges · 1 source tracked
The Maker Tool market, encompassing 3D printers, laser engravers, CNC machines, and UV printers, is experiencing significant growth, with projections indicating a global market size of $36.2 billion by 2033. Leading com…
-
StateFlow framework enables iterative 3D previsualization · 2 sources tracked
StateFlow is a new framework designed to enhance previsualization for creative industries like film, gaming, and architecture. It introduces a persistent, editable 3D world state that allows for iterative refinement of …
-
3D AI Chips: Diverse Architectures, Unified Goal of Enhanced Performance
The AI chip market is experiencing a surge in interest for 3D stacking technologies, driven by the increasing demands of AI agents for greater memory capacity, bandwidth, and data efficiency. While companies are adoptin…
-
AI infrastructure sees soaring B300 prices, new chip architectures, and executive departures
The AI infrastructure market is experiencing significant shifts, with soaring prices for key components like the NVIDIA B300, which has tripled in value over six months. This scarcity is leading to concerns about inform…
-
New SceneActBench benchmark evaluates VLM agents' 3D interaction capabilities
Researchers have introduced SceneActBench, a new benchmark designed to evaluate the capabilities of vision-language model (VLM) agents in interacting with and manipulating 3D environments. Unlike previous benchmarks tha…