RealEstate10K
PulseAugur coverage of RealEstate10K — every cluster mentioning RealEstate10K across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New SPAR3S model generates 3D scenes from sparse multi-view images
Researchers have developed SPAR3S, a novel sparse auto-regressive model for generating complete 3D scenes from limited multi-view images. This method operates in a compact, voxel-aligned 3D latent space, representing on…
-
New methods enhance novel view synthesis stability and quality · 2 sources tracked
Researchers have developed new methods to improve the stability and quality of novel view synthesis, a technique that generates new perspectives of a scene from a limited set of input images. One approach, CamTrol++, fo…
-
ReconSplat model enhances 3D scene reconstruction beyond observed views
Researchers have developed ReconSplat, a novel feed-forward model designed for 3D scene reconstruction that aims to improve the generation of plausible views for unobserved regions while maintaining geometric consistenc…
-
GenRec model separates reconstruction and generation for novel view synthesis
Researchers have introduced GenRec, a novel multi-view flow matching model designed to improve generative novel view synthesis. This model explicitly separates reconstruction from generation, addressing limitations in e…
-
SplatGuide uses 3D Gaussians for pose-free novel view synthesis · arXiv
Researchers have introduced SplatGuide, a novel method for generating photorealistic novel views from unposed images. This approach leverages geometric priors from 3D Gaussians, reusing a single 3D Gaussian Splatting (3…
-
SubSplat method enhances high-resolution 3D rendering efficiency
Researchers have developed SubSplat, a novel method for high-resolution pixel-aligned Gaussian splatting that addresses the computational cost trade-off. SubSplat introduces a Sub-pixel Gaussian Reparameterizer (SPGR) t…
-
New method enhances 3D scene generation using latent space flow matching
Researchers have developed a new method called Latent Riemannian Flow Matching to improve 3D scene generation using geometric foundation models. This technique operates within the latent space of models like the Visual …
-
New benchmarks and frameworks advance video world modeling
Researchers have introduced "ImageTime," a new benchmark designed to evaluate how well image generation models can understand and represent temporal changes. This benchmark assesses spatiotemporal consistency by requiri…
-
EPiC framework enhances video camera control learning
Researchers have developed EPiC, a new framework for efficient video camera control learning. EPiC constructs highly precise anchor videos without requiring camera pose or point cloud estimation, by masking source video…
-
$h$-control method enhances training-free video camera control
Researchers have introduced "$h$-control," a novel method for training-free camera control in video generation models. This approach enhances existing flow-matching techniques by incorporating block-conditional pseudo-G…
-
Pixel-to-4D generates videos from single images with dynamic 3D Gaussians
Researchers have developed a new framework for generating videos from single images, offering enhanced control over camera movement. This method utilizes dynamic 3D Gaussian representations to model scenes and sample ob…