When people think of machine learning, headstones probably don’t come to mind. But in one of the more profound applications of generative AI, we built a system that uses machine learning to design custom memorial products—yes, including gravestones—tailored to personal preferences and cultural sensitivities.
This wasn’t just a quirky ML experiment. It was a full-stack application of generative models, natural language processing, and human-in-the-loop systems, all to address a highly sensitive and deeply human need: commemorating a life.
The Problem: Designing With Dignity
Dizajn spomenika je umjetnost i tradicija.Obitelji žele nešto osobno, poštovanje i često simbolično.Proces dizajna je spor, emocionalno opterećujući i ograničen materijalima, propisima o groblju i vjerskim ili kulturnim tradicijama.
We set out to build something that could help—not replace—designers: a headstone generator that could produce realistic, meaningful design options based on prior data and customer preferences.
You can try it here: Headstonesdesigner.com/generator Svijet(svi podaci o osposobljavanju dolaze iz live mjesta -https://headstonesdesigner.com/)
Korak 1: Razumijevanje domene
Before we touched TensorFlow or wrote a single line of code, we immersed ourselves in the world of memorial art. We studied:
- Traditional and contemporary styles
- Religious and cultural norms
- Material constraints (granite, marble, etc.)
- Pravila groblja, kao što je max širina spomenika po parceli
Dizajniranje AI-a za osjetljivu domenu kao što je ova zahtijeva duboko poštovanje i nijanse.
Korak 2: Izgradnja baze podataka
We pulled together a surprisingly diverse dataset:
- Tisuće dizajnerskih fotografija
- CAD datoteke postojećih glavnih kamenaca
- Customer preference history
- Tekst iz upisa
- Cemetery dimensional standards
All of this needed to be cleaned, normalized, and vectorized. Texts were embedded using models like BERT. Images were preprocessed and augmented. This wasn’t just about throwing data into a model—it was about making it learnable.
Korak 3: Model arhitekture i obuka
We tested a few model types in parallel:
- StyleGAN2: Za stvaranje visokokvalitetnih, stiliziranih slika spomenika
- VAEs (Variational Autoencoders): For interpolating between design styles and enabling user-controlled variations
- Transformeri (GPT): Za stvaranje natpisa koji su se osjećali osobnim, relevantnim i poštovanim
Posebno je kompliciran dio bio osiguravanje da se tekst i vizualni materijali podudaraju.Gotički kamen ne bi trebao imati Comic Sans natpise.
To smo riješili s:
- Multi-modal training: Combining vision and language models (like CLIP) to assess alignment
- Conditional GANs: Using the text as input to guide visual generation
Step 4: Managing the Unknowns
We had plenty of “AI gone weird” moments.
- Some early outputs looked more like modernist sculpture than memorials.
- Prijenos stila ponekad je prešao kulturne linije na neugodne načine.
- GPT occasionally generated tone-deaf epitaphs.
Kako bi se to ublažilo, izgradili smo povratne informacije od ljudi u krugu. Dizajneri i kulturni savjetnici pregledali su rezultate i označili probleme.
We also used techniques like style discriminators in GANs to enforce constraints and post-generation filters to validate text content.
Korak 5: Procjena i rezultati
We didn’t just eyeball the results. Evaluation was multi-pronged:
- FID ocjene za imaginarni realizam
- BLEU ocjene i ljudska procjena točnosti teksta
- Istraživanja korisnika i stručnjaci za estetsku i kulturnu vjernost
The final result? A system that could generate emotionally resonant, visually accurate, and context-aware headstone designs.
Ovdje možete komunicirati s generatorom:Headstonesdesigner.com/generator Svijet
Lessons Learned
Some takeaways:
- Cultural context isn’t an edge case—it's the core requirement in sensitive design domains.
- Generative AI is powerful, but without constraints, it easily drifts into uncanny or inappropriate territory.
- Human feedback isn't just helpful; it's mandatory.
The Future
We’re exploring how this tech could extend into other domains: wedding invitation design, personalized awards, commemorative art, and more. Anywhere design is personal and high-stakes, there's an opportunity to blend generative ML with human care.
