Abstract

Film emulation reproduces the look of an analog film stock on a new digital photograph. We target its open-set form—matching the look of any reference film frame from a single example—with a 3D lookup table (LUT) predicted from that reference. Real-time image enhancement has converged on one recipe for the color step: predict per-image weights over a fixed bank of 3D LUTs and blend them. We show this construction is a gated mixture of experts—specialist tables chosen per image by a learned gate—and it inherits the corresponding failure: trained end-to-end against reconstruction, the gate collapses onto a single expert, so a bank of K LUTs delivers the capacity of one. In a five-LUT model, one table draws 95% of the weight and four sit idle (mean gate entropy 0.16 of a possible 1.61). An entropy term, the enhancement-setting analogue of mixture-of-experts load balancing, restores utilization and recovers about 1 dB PSNR. The deeper constraint survives the fix: a fixed LUT basis is closed-set, freezing the achievable looks at training time. We therefore discard the basis and predict a single 3D LUT as a residual from a reference image (StyleLUTNet), trained by self-supervision on procedurally generated color transforms. The conditional design removes the gate and generalizes open-set to arbitrary, unseen film stocks without paired data or retraining. Around this color backbone we build Deep Analog, a film-emulation pipeline that adds histogram-based tone matching and a physics-informed optical renderer—multi-scale grain and per-channel halation driven by parameters an inverse network regresses from the reference. On 350 self-supervised pairs the color stage reaches 22.05 dB PSNR / 0.925 SSIM and the full pipeline 21.72 dB / 0.923; the color path runs in 5.2 ms at 1080p (192 FPS) and exports a portable .cube LUT for Photoshop, DaVinci Resolve, and Lightroom. A second degeneracy in conditional LUT training—residual-scale collapse—shares the root cause and yields a general operating principle: auxiliary regularization must stay subordinate to reconstruction.

Publication Date

8-2026

Document Type

Thesis

Student Type

Graduate

Degree Name

Computer Science (MS)

Department, Program, or Center

Computer Science, Department of

College

Golisano College of Computing and Information Sciences

Advisor

Qiuxiao Chen

Advisor/Committee Member

Michael Murdoch

Advisor/Committee Member

Joe Geigel

Campus

RIT – Main Campus

Download

Share

COinS