Benchmarking Trellis 2: GLB Output Quality Across Six Input Types
Description and use
Testing Trellis 2 revealed that phone photos and concept art yield the best GLB outputs for Unity, offering a quick shortcut to studio shots.
Problem
Game developers need to know how well Trellis 2 converts various image inputs into usable GLB assets for Unity.
Outcome
Trellis 2 reliably produces clean GLBs from studio shots and performs best with phone photos and concept art, offering a viable shortcut to studio‑shot workflows for game devs.
Workflow
- Select six input types: studio shot, phone photo, concept art, low‑poly screenshot, sketch, collage.
- Upload each image to Trellis 2 via browser.
- Export GLB and import into Unity.
- Evaluate silhouette accuracy, topology sanity, texture coherence, and import time.
- Score each input on a 1–10 scale.
- Compile results and derive practical takeaways.
- Publish findings for dev community.
Evidence
Six input types, one subject each, all run through Trellis 2 in the browser (it's cloud‑based, so no local GPU variables in the mix).
Studio shot: Flawless, as advertised. Clean topology, 2/2 import.
Phone photo: The surprise — the model ignored the desk clutter impressively well. Silhouette true, some texture hallucination on the back side (which no single-photo method can know). Usable for background props with 10 minutes of cleanup. 8/10.
Concept art: This is where SLAT's learned-prior approach shines. The stylized painting produced a stylized mesh that matched the art direction — something photogrammetry literally cannot do (it would need the object to exist). For stylized games, this is the killer input type. 9/10.
Low-poly screenshot: The "remaster" workflow works: feed an old asset's screenshot, get a modernized mesh. Ironically the output sometimes inherited the input's angularity — garbage in, slightly better garbage out. Still a legitimate retro-asset pipeline. 7/10.
Sketch: The wildest card. A pencil sketch produced a coherent, charming "concept-model" — wrong in the ways your drawing was wrong. As a tool for pre‑visualizing your own sketch in 3D, genuinely useful. As production asset, no. 6/10 for production, 9/10 for ideation.
Collage: Mixed views in one image confused the reconstruction more than helped. Front-only input beat my hacky collage. Don't do this — if you want multi-view, that's what dedicated modes are for. 4/10.
Best input: clean silhouette, one dominant view, honest lighting. Backgrounds matter less than I feared.
For game devs, the sweet spot is inputs 2 and 3: real‑world reference photos and concept art. That covers about 80% of "I need a prop by Friday" situations.
- Record date: 2026-10-27T00:11:16.649Z