Papers4

All Beginner Intermediate Advanced

All Sources arXiv

#PSNR SSIM LPIPS

PLANING: A Loosely Coupled Triangle-Gaussian Framework for Streaming 3D Reconstruction

Intermediate

Changjian Jiang, Kerui Ren et al.Jan 29arXiv

PLANING is a new way to build 3D worlds from a moving single camera by combining two kinds of pieces: sharp triangles for shape and soft Gaussians for looks.

#Streaming 3D Reconstruction#Triangle Primitives#Neural Gaussians

Implicit Neural Representation Facilitates Unified Universal Vision Encoding

Intermediate

Matthew Gwilliam, Xiao Wang et al.Jan 20arXiv

This paper introduces HUVR, a single vision model that can both recognize what’s in an image and reconstruct or generate images from tiny codes.

#Implicit Neural Representation#Hyper-Networks#Vision Transformer

Pretraining Frame Preservation in Autoregressive Video Memory Compression

Intermediate

Lvmin Zhang, Shengqu Cai et al.Dec 29arXiv

The paper teaches a video model to squeeze long video history into a tiny memory while still keeping sharp details in single frames.

#autoregressive video generation#video memory compression#frame retrieval pretraining

StereoPilot: Learning Unified and Efficient Stereo Conversion via Generative Priors

Intermediate

Guibao Shen, Yihua Du et al.Dec 18arXiv

StereoPilot is a new AI that turns regular 2D videos into 3D (stereo) videos quickly and with high quality.

#stereo video conversion#monocular-to-stereo#depth ambiguity