AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Learning Compact 3D Representations from Feed-Forward Novel View Synthesis

Honggyu An, Jaewoo Jung, Mungyeom Kim, Chaehyun Kim, Minkyeong Jeon, Jisang Han, Kazumi Fukuda, Takuya Narihira, Hyunah Ko, Junsu Kim, Sunghwan Hong, Yuki Mitsufuji and 1 more

CVPR 2026 · 3D vision › Gaussian splatting · approach: Sparse-view Gaussian splatting for novel views

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add progressive low-pass filter during training (ablation)C3G-G · RealEstate10KPSNR: N/A -> 22.387Improved
unfreeze the visual encoder (ablation)C3G-G · RealEstate10KPSNR: 18.441 -> 22.387Improved
increase learnable query tokens from 2048 to 4096 (ablation)C3G-G · RealEstate10KPSNR: 20.625 -> 19.012Hurt
increase learnable query tokens from 256 to 2048 (ablation)C3G-G · RealEstate10KPSNR: 19.713 -> 20.625Improved
use VGGT instead of DINOv3 as visual encoder (ablation)C3G-G · RealEstate10KPSNR: 20.292 -> 22.387Improved
use C3G-F feature decoder instead of none (ablation)C3G-F · ScanNet (LSeg)mIoU: 0.193 -> 0.513Improved
add autoencoder feature compression to C3G-F (ablation)C3G-F · ScanNet (LSeg)mIoU: 0.513 -> 0.512No clear effect
add C3G feature lifting on top of VGGT-Tracking featuresVGGT-Tracking · ScanNet correspondenceaverage PCK@10px: 34.0 -> 68.1Improved
add AnyUp upsampler to DINOv3-Large featuresDINOv3-Large · ScanNet correspondenceaverage PCK@10px: 37.7 -> 31.8Hurt
add AnyUp upsampler to DINOv2-Base featuresDINOv2-Base · ScanNet correspondenceaverage PCK@10px: 23.8 -> 23.9No clear effect
use more input views with per-pixel Gaussians (VGGT+NoPo)VGGT+NoPo · RealEstate10KPSNR: 21.262 -> 21.190Hurt
add test-time optimizationC3G · RealEstate10K 36 viewsPSNR: 23.812 -> 30.250Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy