AI Paper Atlas

Research map / 3D vision / Face and shape

Modulated Periodic Activations for Generalizable Local Functional Representations

Ishit Mehta, Michaël Gharbi, Connelly Barnes, Eli Shechtman, Ravi Ramamoorthi, Manmohan Chandraker

ICCV 2021 · 3D vision › Face and shape · approach: Implicit neural representation design · 134 citations (OpenAlex)

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
use modulation network instead of concatenating latent code with coordinates in a sine MLP (ablation)sine synthesis network · CelebA 32x32PSNR: 20.15 -> 29.42Improved
use modulation network instead of concatenation on a harder dataset (ablation)sine synthesis network · CIFAR-10PSNR: 25.73Improved
use modulator instead of hypernetwork conditioning (ablation)SIREN synthesis network · CelebA 32x32PSNR: 26.70 -> 29.42Improved
use sine activations instead of ReLU in the synthesis network (local model) (ablation)local functional representation · Div2K 2K imagesPSNR: 34.73 -> 38.03Improved
use sine activations instead of ReLU in the synthesis network (local model) (ablation)local functional representation · ThreedScans Scene AChamfer distance (x1e-5): 1.37 -> 1.32Improved
use sine activations instead of ReLU in the synthesis network (local model) (ablation)local functional representation · ThreedScans Scene BChamfer distance (x1e-5): 8.37 -> 2.40Improved
add local tiling with pretrained generalizable model instead of per-image global overfittinghigh-resolution image encoding · Div2K 2K imagesPSNR: 28.48 -> 38.03Improved
add local tiling with pretrained model to video encodinghigh-res video encoding · 6 pexels videos 640x256PSNR: 25.19 -> 25.21No clear effect
use FFN sigma=10 instead of sigma=1 (ablation)FFN conditional baseline · CelebA 32x32 (2x resolution)PSNR: 24.37 -> 4.22Hurt
use larger convolutional encoder (17M params) (ablation)Ours · CelebA 32x32PSNR: 29.42 -> 29.64Improved
use modulator network instead of DeepSDF ReLU concatenationconditional SDF network · ShapeNet carsChamfer distance (median): 0.00284 -> 0.00230Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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