Seedlabs

RadianceCrop: Aesthetic Auto-Crop on HDR Radiance Maps for Product Photography

An API for product photos that merges multiple exposures and crops the image to keep the best lighting.

An API for e-commerce photo pipelines that merges bracketed exposures into an HDR radiance map, then automatically crops and retargets the image to meet composition guidelines — so specular highlights and shadow detail in salient regions survive the crop and final tone map.

Computer ScienceImage Enhancement Techniques
Aesthetic Perception and AnalysisE-commerce product photography automation

Concept

An API/plugin for catalog and marketplace photo pipelines that produces automatically composed, correctly cropped image variants from a single bracketed capture. Today's automatic composition tools operate on finished 8-bit images, where specular highlights are already clipped and shadows already crushed. This product runs the aesthetic step upstream, on the scene-referred radiance map: (1) merge the exposure bracket into an HDR radiance map; (2) detect salient regions; (3) apply a crop-and-retarget operator to reposition them according to established composition guidelines; (4) tone-map the result to each required aspect ratio (marketplace square, banner, thumbnail). Because the operator works on full luminance data, salient highlights — the sparkle on a watch bezel, the glint on a ring — survive both the retarget and the final tone map. For glossy products this is the difference between a flat crop and a sale-ready one.

Why now

Paper [2] demonstrates that composition aesthetics can be evaluated computationally against well-grounded guidelines and modified with a compound crop-and-retarget operator that changes the relative position of salient regions in the image. Paper [0] demonstrates that HDR imaging has matured into a practical, codified discipline — with detailed equations and code — that captures the full dynamic range of visible light and unifies photographic capture with graphics-style processing. Together they make a new combination practical: running the aesthetic operator on scene-referred radiance rather than clipped camera output, which existing composition tools cannot do.

Written by GLM · 11 Oct 2026

AI assessment

Backed by 2 papers84

A specialized API that optimizes product photo composition on HDR data to preserve high-value specular highlights, offering a clear technical edge over standard 8-bit cropping tools.

Evidence strength
4/5
The idea logically synthesizes a well-established HDR framework [2] with a specific computational composition operator [1], though it assumes the operator's effectiveness translates perfectly from 2D images to radiance maps.
Market pull
4/5
High-end e-commerce (jewelry, watches) has a strong incentive to automate the tedious process of bracketed capture and manual cropping while maintaining 'luxury' visual cues like glints.
Novelty & moat
3/5
While ViewThrough uses HDR for real estate, applying it specifically to aesthetic composition and retargeting for product catalogs is a distinct and valuable pivot.
Feasibility
5/5
Both HDR merging and crop-and-retarget operators are mathematically codified and implementable as a software pipeline without requiring new hardware or breakthroughs.
Wedge clarity
5/5
The focus on 'specular highlights in salient regions' for glossy products provides a sharp, high-value entry point into the e-commerce pipeline.
Simplicity / focus
5/5
The product is a single, focused API with a clear input (bracketed photos) and output (optimized, tone-mapped crops).

Scored by Gemma 4 · 11 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • Chrono24company

    The global watch marketplace depends on listing photos where clipped highlights on metal and poor crops depress buyer interest; automated HDR capture plus guideline-compliant crops raises listing quality across its huge inventory.

  • Brilliant Earthcompany

    Jewelry e-commerce product photography lives or dies on specular sparkle; a radiance-map pipeline preserves the highlight detail that single exposures clip, and automated composed variants speed catalog production.

  • Adobe Inc.company

    Could integrate radiance-map-aware auto-crop into Lightroom/Photoshop, extending its existing HDR-merge and auto-crop features with a composition-guided step.

  • Shopifycompany

    Merchant catalogs need consistent, well-composed imagery across many aspect ratios; an API that outputs composed variants from one bracket reduces listing friction for merchants.

Research it builds on

  1. Optimizing Photo Composition
    Ligang Liu, Renjie Chen, Lior Wolf et al. · 2010 · 258 citations
    All ideas from this paper →
  2. Advanced high dynamic range imaging : theory and practice
    Francesco Banterle, Alessandro Artusi, Kurt Debattista et al. · 2026 · 217 citations
    All ideas from this paper →
Leave feedback
feasibility