modelbeast/server/operators/colmap_poses/run.py
MODELBEAST 605b1ae347 Phase 1 + framework: settings/secrets, queue lanes, job mgmt, inbox, 8 new operators
Framework:
- server/settings.py: key/value settings + secrets, env-injected into operator
  subprocesses, secret values masked in API and redacted from job logs
- runner: gpu/cpu/net concurrency lanes, job cancel/retry/delete, multi-input,
  graceful 'not installed' error when a tool venv is missing
- db: settings table, asset_ids column (migrated), MODELBEAST_DATA test override
- main: settings + job-action endpoints, inbox watch folder auto-ingest
- store: operators can tag output asset kind (splat, colmap_dataset)

Operators (11 total):
- fal_trellis/trellis2/hunyuan3d/rodin via shared _lib/fal_common.py (verified
  params + endpoint ids; recursive result-URL extractor handles per-endpoint keys)
- sf3d, trellis_mac: local MPS image-to-3D, installed with Metal kernels built,
  gated on owner HuggingFace auth
- colmap_poses (COLMAP 4.x + GLOMAP global mapper), brush_train (native Metal 3DGS)
- Scan pipeline validated end-to-end through the UI: frames -> colmap (48/48
  registered, 0.6px) -> brush -> splat.ply -> in-app SplatViewer

Frontend:
- Settings modal, operator gating (lock + disabled run when requires_env unmet),
  job cancel/retry/delete, Compare grid (multi-select side-by-side viewers),
  SplatViewer (gaussian-splats-3d, Ply format forced for extensionless URLs)

Tooling: scripts/install_{colmap,brush,sf3d,trellis_mac}.sh; vendor/ + venvs/
gitignored; tests/smoke.sh (12 checks passing); BENCHMARKS.md

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-12 21:42:27 +10:00

77 lines
2.7 KiB
Python

import argparse
import json
import shutil
import subprocess
import sys
from pathlib import Path
ap = argparse.ArgumentParser()
ap.add_argument("--input", action="append", default=[])
ap.add_argument("--outdir", required=True)
ap.add_argument("--params", default="{}")
a = ap.parse_args()
p = json.loads(a.params)
if not a.input:
print("ERROR: no input frames folder")
sys.exit(1)
src = Path(a.input[0])
if not src.is_dir():
print(f"ERROR: input must be a frames folder, got {src}")
sys.exit(1)
outdir = Path(a.outdir)
dataset = outdir / "colmap_dataset"
images = dataset / "images"
images.mkdir(parents=True, exist_ok=True)
frames = sorted([f for f in src.iterdir() if f.suffix.lower() in (".jpg", ".jpeg", ".png")])
if len(frames) < 3:
print(f"ERROR: need >=3 frames, found {len(frames)}")
sys.exit(1)
for f in frames:
shutil.copy2(f, images / f.name)
print(f"copied {len(frames)} frames", flush=True)
db_path = dataset / "database.db"
sparse = dataset / "sparse"
sparse.mkdir(exist_ok=True)
def run(*cmd):
print("+", " ".join(str(c) for c in cmd), flush=True)
r = subprocess.run([str(c) for c in cmd], stdout=sys.stdout, stderr=subprocess.STDOUT)
if r.returncode != 0:
print(f"ERROR: command failed ({r.returncode})")
sys.exit(r.returncode)
run("colmap", "feature_extractor", "--database_path", db_path, "--image_path", images,
"--ImageReader.single_camera", "1" if p.get("single_camera", True) else "0",
"--ImageReader.camera_model", p.get("camera_model", "OPENCV"),
"--FeatureExtraction.use_gpu", "0")
matcher = "sequential_matcher" if p.get("matcher", "sequential") == "sequential" else "exhaustive_matcher"
run("colmap", matcher, "--database_path", db_path, "--FeatureMatching.use_gpu", "0")
if p.get("mapper", "global") == "global":
run("colmap", "global_mapper", "--database_path", db_path,
"--image_path", images, "--output_path", sparse)
else:
run("colmap", "mapper", "--database_path", db_path,
"--image_path", images, "--output_path", sparse)
# report registered images
recon = sparse / "0"
n_reg = 0
if (recon / "images.bin").exists():
n_reg = (recon / "images.bin").stat().st_size # crude presence signal
print(f"reconstruction at {recon}, images.bin present: {(recon / 'images.bin').exists()}", flush=True)
if not (recon / "images.bin").exists() and not (recon / "images.txt").exists():
print("ERROR: COLMAP produced no reconstruction (too few features/matches?)")
sys.exit(1)
(outdir / "result.json").write_text(json.dumps({"outputs": [
{"path": str(dataset), "name": f"{src.name}_colmap",
"meta": {"kind": "colmap_dataset", "frames": len(frames)}}]}))
print("done", flush=True)