Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

AdaptiveSplat:Texture Aware Controllable 3D Gaussian Allocation for Feed-Forward Reconstruction

About

Current feed-forward 3D reconstruction methods predict pixel aligned Gaussian primitives, resulting in highly redundant representations. A natural solution is to prune the redundant Gaussians, but naive pruning introduces severe artifacts and often requires inference time fine-tuning, breaking the feed-forward paradigm. Based on previous works, high frequency regions require more Gaussian primitives, while low frequency regions can be represented with significantly fewer primitives. Motivated by this, we propose a novel approach to explicitly control the number of Gaussians by leveraging local texture information. Our approach achieves this through three key components: (1) texture estimation to capture spatial variation in scene detail, (2) texture-aware pruning that removes redundant Gaussians from low frequency regions, and (3) an adaptive Gaussian head that predicts the modified attributes of the retained primitives without breaking the feed-forward paradigm. Experiments on RE10K, ACID, DL3DV, Tanks and Temples, and DTU demonstrate the effectiveness of our approach, while ablation studies validate the contributions of its key components.

Badrinath Singhal, Srihari K G, Sreehari Iyer, Ankit Dhiman, Venkatesh Babu Radhakrishnan• 2026

Related benchmarks

TaskDatasetResultRank
Novel View SynthesisRE10K
SSIM74
345
Novel View SynthesisACID
PSNR22.55
175
Novel View SynthesisDTU
PSNR15.23
154
Novel View SynthesisDTU (test)
PSNR16.19
119
Novel View SynthesisACID (test)
PSNR22.55
113
Novel View SynthesisDL3DV 6view
PSNR20.448
46
Novel View SynthesisDL3DV 32 views
PSNR17.677
37
View SynthesisRe10K (test)
PSNR20.74
30
Novel View SynthesisDL3DV 9 views
PSNR19.678
24
Novel View SynthesisDL3DV 16 Views
PSNR19.125
24
Showing 10 of 15 rows

Other info

Follow for update