← back to test matrix

Test 4.2
BLOCKED — needs pod

Can an open-vocabulary detector find the boom and battens, zero-shot?

BLOCKED — needs a GPU pod: a zero-shot screen with MM-Grounding-DINO could not run on Nate's Mac without risking a repeat of a prior out-of-memory crash. No boxes were drawn, no hit rate exists yet.
What this test was checking

The idea: point an off-the-shelf detector at our sail photos with the plain-English labels "boom" and "batten pocket" — no training, no fine-tuning — and see if it can box those parts on its own. If it can, that's a free signal to help SAM3 draw sail outlines or catch a stripe line that clearly isn't on the sail. This was meant to be a quick 10-photo eyeball check, not a real build.

Why it's blocked

A previous attempt at this exact test caused the detector's Python process to balloon to 6.5GB of memory and lock up the machine. This session checked before touching anything: none of the required software (mmdet, mmcv, mmengine) is installed, no model weights are cached anywhere on disk, and the piece that caused the crash last time — the model itself running in memory — is a known 6.5GB-plus load, not a small download. Rather than risk repeating that crash, this session stopped and confirmed the safe path is a cloud GPU pod, the same kind already used for other model spikes in this project.

CheckResult
Model weights cached locallyNo — nothing found anywhere on disk
Required software installedNo — mmdet/mmcv/mmengine/transformers all missing from the inference environment
Forward passes run0
Photos screened0 of the planned ~10
Smallest checkpoint found (HF transformers variant, "tiny")~657 MB on disk
Checkpoint for the actual MM-Grounding-DINO-Tiny model this test is named for~916 MB on disk (needs mmdet/mmcv, not installed)
Known runtime memory need (from the prior crash on this same test)6.5 GB — this is the real blocker, not the download size
Machine put at risk this sessionNo — only checked file sizes over the network, no model was ever loaded
What a pod run would need

Any cloud GPU box (nothing exotic — the same class used for prior SAM3 spikes here), the detector loaded with the "boom" / "batten pocket" text prompts, and about 10–20 photos from the frozen holdout set streamed one at a time. Under an hour of GPU time. Someone would then eyeball whether the boxes land on the real boom and battens and report a hit rate — that's the actual test, still not run.

Source: model-mmgdino.json.

raw data (JSON) →