C1/C4: Cap blur-score computation at 1440 px before calling assess_quality so scores are always on the same Laplacian scale as the embedding path (which operates on Immich preview thumbnails). Also converts the image to RGB before scoring and stores 0.0 on assess_quality failure so files uploaded without a score remain eligible for future quality replacement instead of occupying a slot permanently. C2: Fall back to the tracker's mapped-filename set as the pre-upload baseline when the Frigate GET /api/faces endpoint is unreachable at upload start. Previously, uploads that succeeded during a partial API outage were never mapped in frigate_files, leaving get_tracked_frigate_file_count permanently under-counting those files and allowing Frigate to exceed MAX_AUTO_IMAGES over time. C3: Track min_quality_score_for_slot when a quality-replacement delete succeeds but the subsequent upload fails. This ensures the freed slot can only be filled by a candidate that beats the deleted file's score, not just the next file in iteration order (which could be lower quality than what was deleted). Add get_tracked_frigate_filenames() to upload_tracker and expand tracker tests to cover the new function and exclude-parameter behaviour. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
winnow
Docs: Setup · Troubleshooting · FAQ
winnow pulls photos from your Immich library, selects the most diverse and highest-quality subset using AI embeddings, and delivers them as training data for Frigate's face recognition and object classification models.
Frigate's face recognition is only as good as its training data — and the key quality metric is diversity, not volume. A hundred photos from the same week teach the model one lighting condition. What you need is a spread: different years, different angles, different lighting, different contexts. Your photo library already has that data. winnow finds and delivers the right subset automatically.
How It Works
Immich library
│
▼
1. Fetch all assets tagged with this person
│
▼
2. Filter by recency (YEARS_FILTER) and skip already-uploaded
and rejected assets (persistent tracker in CACHE_DIR)
│
▼
3. Quality filter — download preview thumbnails and reject:
• Blurry images (Laplacian variance)
• Grayscale / infrared (channel similarity)
• Over- or underexposed
• Low detection confidence
• Face crops below minimum pixel size
│
▼
4. Compute embeddings from the same preview thumbnails
• Faces → InsightFace (ArcFace / Buffalo_L) → 512-dim vector
• Objects → SigLIP (Vision Transformer) → 768-dim vector
│
▼
5. Diversity selection
• K-Medoids clustering → one representative per natural group
• Farthest Point Sampling → fill remaining slots with maximally spread picks
• Hard example weighting — unusual angles and low-confidence detections
are biased toward selection, since those are where models tend to fail
• Auto mode: stops when similarity to the existing set exceeds a threshold
(20 % of median pairwise distance for faces, 10 % for objects)
│
▼
6. Download full-resolution originals from Immich
│
▼
7. Crop and process
• Face mode: EXIF-corrected, landmark-aligned 112×112 crop (ArcFace format)
• Object mode: YOLOv9c detection → one crop per matched instance
│
▼
8. Deliver
• Face mode: upload crops to Frigate's face registration API
↳ below MAX_AUTO_IMAGES — upload freely
↳ at cap + QUALITY_REPLACEMENT=true — swap the lowest-scoring tracked
image if the new candidate scores higher; manually added files are
never touched
↳ at cap + QUALITY_REPLACEMENT=false — skip this person
• Object mode: save crops to disk → place into your Frigate data directory
Uploaded and rejected asset IDs are persisted across runs. The same image is never processed twice; Frigate rejections are permanently skipped unless RETRY_REJECTED=true.
Modes
Face mode (default) — extracts face crops using Immich's bounding box metadata, applies EXIF orientation correction, and aligns them to ArcFace's standard 112×112 format using 5-point facial landmarks. Crops are uploaded directly to Frigate's face registration API.
Object mode — runs each full-resolution image through YOLOv9c to detect instances of a target class (dog, cat, car, etc.), crops each detection, and saves it to the output directory. Frigate has no API for uploading object training data; place the crops into your Frigate data directory manually.
Running in Docker
Image Tags
| Tag | Arch | Acceleration |
|---|---|---|
:latest |
amd64 + arm64 | NVIDIA CUDA 13.3 (amd64) · requires NVIDIA Container Toolkit |
:rocm |
amd64 | AMD ROCm · pass /dev/kfd + /dev/dri |
:intel |
amd64 | Intel Arc / iGPU via OpenVINO · pass /dev/dri, set OPENVINO_DEVICE=GPU |
:cpu |
amd64 + arm64 | CPU only · ~2 GB smaller · no GPU required |
Quick Start
NVIDIA:
services:
winnow:
image: ghcr.io/sudolulo/winnow:latest
environment:
- IMMICH_URL=http://192.168.1.10:2283
- API_KEY=your-immich-api-key
- FRIGATE_URL=http://192.168.1.10:5000
- CRON_SCHEDULE=0 3 * * 0
volumes:
- /path/to/models:/models
- /path/to/cache:/app/.if_cache
- /path/to/output:/app/frigate_train
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
AMD (:rocm): use image: ghcr.io/sudolulo/winnow:rocm and replace the deploy: block with:
devices:
- /dev/kfd
- /dev/dri
group_add:
- video
- render
Intel (:intel): use image: ghcr.io/sudolulo/winnow:intel and replace the deploy: block with:
devices:
- /dev/dri
group_add:
- render
environment:
- OPENVINO_DEVICE=GPU # omit to run OpenVINO inference on CPU (default)
CPU (:cpu): use image: ghcr.io/sudolulo/winnow:cpu, remove the deploy: block, and add mem_limit: 2g to prevent OOM on large libraries.
See compose.yml for the full annotated example with all options.
Scheduling
CRON_SCHEDULE controls container lifetime:
CRON_SCHEDULE value |
Behaviour |
|---|---|
| (unset) | Run once on startup, then exit |
| (empty string) | Stay alive, run nothing — trigger manually with docker exec -it winnow winnow |
| Cron expression | Run on startup, then repeat on schedule |
In scheduled mode the process (and loaded models) stays resident between runs. The first run after a fresh install downloads the embedding models (~1–2 GB); subsequent runs use the cached models from the mounted volume.
Environment Variables
Connection
| Variable | Default | Description |
|---|---|---|
IMMICH_URL |
(required) | Full URL to your Immich instance |
API_KEY |
(required) | Immich API key |
FRIGATE_URL |
(unset) | Frigate URL — required for face upload; omit to skip |
Mode & Strategy
| Variable | Default | Description |
|---|---|---|
TRAINING_MODE |
face |
face — upload crops to Frigate; object — save crops to disk |
STRATEGY |
auto |
auto (embedding-based adaptive), standard (30 images), broad (100 images) |
LIMIT |
(unset) | Exact image count — overrides STRATEGY |
OBJECT_CLASS |
dog |
Target class for object mode (any YOLO class: dog, cat, car, etc.) |
AUTO_MODE |
(auto) | Force non-interactive mode in a terminal; auto-detected otherwise |
VERBOSE |
false |
Enable DEBUG-level console output (log file is always DEBUG) |
People Filtering
| Variable | Default | Description |
|---|---|---|
ONLY_PEOPLE |
(unset) | Comma-separated whitelist — process only these people |
SKIP_PEOPLE |
(unset) | Comma-separated list — skip these people |
MIN_FACE_COUNT |
0 |
Skip people with fewer than N tagged assets in Immich |
YEARS_FILTER |
10 |
Ignore images older than N years |
Image Quality
| Variable | Default | Description |
|---|---|---|
MIN_FACE_WIDTH |
50 |
Minimum face crop width in pixels |
FACE_MARGIN |
0.15 |
Padding around bounding box crop (fraction of face size) |
ENABLE_FACE_ALIGNMENT |
true |
Align to ArcFace 112×112 format using facial landmarks |
USE_FULL_RESOLUTION |
true |
Download full-resolution originals rather than preview thumbnails |
MIN_CONFIDENCE |
0.7 |
Minimum Immich face detection confidence |
BLUR_THRESHOLD |
100.0 |
Laplacian variance threshold — lower accepts more blur |
MAX_AUTO_IMAGES |
80 |
Maximum training images per person in Frigate |
QUALITY_REPLACEMENT |
true |
When at cap, swap the lowest-scoring tracked image for a better candidate. Never touches manually added Frigate files. Set false to skip people already at cap |
GPU & Models
| Variable | Default | Description |
|---|---|---|
FORCE_CPU |
false |
Disable GPU — fall back to CPU for all inference |
OPENVINO_DEVICE |
CPU |
Intel variant only: set GPU to use Arc or iGPU; default runs on CPU |
ENABLE_CACHE |
true |
Cache computed embeddings to disk (speeds up re-runs on the same library) |
CACHE_DIR |
.if_cache |
Path for embedding cache and upload tracker files |
HF_HOME |
(system) | HuggingFace model cache path (SigLIP) |
INSIGHTFACE_HOME |
(system) | InsightFace model cache path (Buffalo_L) |
Output
| Variable | Default | Description |
|---|---|---|
OUTPUT_DIR |
./frigate_train |
Directory for object-mode crops and the winnow.log file. In Docker, set this via the volume mount instead. |
Tracker Overrides (one-shot — remove after use)
| Variable | Default | Description |
|---|---|---|
DRY_RUN |
false |
Preview selection without downloading or uploading |
RETRY_REJECTED |
false |
Re-attempt assets previously rejected by Frigate |
RESET_PERSON |
(unset) | Clear upload and rejection history for one person by name |
Scheduling
| Variable | Default | Description |
|---|---|---|
CRON_SCHEDULE |
(unset) | Unset = run once and exit; empty = stay alive; cron expression = scheduled |
Local Install
git clone https://github.com/sudolulo/winnow.git
cd winnow
uv sync
uv run winnow
Requires Python 3.13+ and uv. An NVIDIA, AMD, or Intel GPU is recommended — CPU mode works but embedding computation is slower.
When run with a terminal attached, winnow starts an interactive session: select which people to process and choose a strategy (auto, standard, broad, or a custom count) per person. Without a TTY — Docker, cron, or AUTO_MODE=true — it processes all people automatically using the configured defaults.
Requirements
- Immich v1.106+
- Frigate v0.16+ (face mode only — object mode has no Frigate dependency)
- GPU recommended: NVIDIA (CUDA), AMD (ROCm), or Intel (Arc / iGPU via OpenVINO)
- Python 3.13+
Getting Help
- GitHub Discussions — questions, setup help, and general discussion
- Wiki — setup guide, troubleshooting, and FAQ
- Issues — bugs and feature requests only
Attribution
Based on if_curator by Sebastian, licensed MIT.