RUN GUIDE — chipoint-2 street-view + general-photo arms

Files

street_gallery/proj_head_gallery_512_{dinov3,so400m256,gopt}.npy — street gallery projection vectors (4.89M OSV-5M street frames, unit-norm 512-d fp16, the street-arm index)
street_gallery/gallery_all_C_so400m256.npy — street gallery coordinates (4.89M x 2)
weights/proj_head_{dinov3,so400m256,gopt}.pt — projection heads (tower -> 512-d)
weights/streetview_k512_joint.pt — street-view reranker arm (K512, 23 features)
photo_gallery/ — unified photo gallery, 14,541,898 rows (MP14 10.9M + MP16-Pro 4.1M): emb_*.npy (512-d fp16 projected), ll_*.npy (lat/lon), ids_*.txt (source ids, row-aligned), manifest.json
weights/photoarm_k768_thr.pt / weights/photoarm_k768_sr.pt / weights/photoarm_k512_thr.pt — general-photo reranker arms (23 features)
weights/ghead_merged_z4/z6/z8/z10.pt, weights/ghead_merged_reg.pt, features/gallery_merged_cells.npz — merged-head feature stack (required for the 23-feature contract)
weights/router_osv_flickr.pt — street/photo router (DINOv3 512-d -> logit;

0 photo, else street)

Load

python run_chipoint.py weights/streetview_k512_joint.pt   # street arm: verify + schema
python run_chipoint.py weights/photoarm_k768_thr.pt       # photo arm: verify + schema

Checkpoint dict: state (SR), thr_state (ThrNet), mu/sd (normalizers), nf, names, thr_km.

Feature contract (nf = 23)

18 base features over the candidate pool (per-encoder sims, margins, ranks, cross-encoder agreement, vote counts, spatial density) + 5 merged-head features from ghead_merged_z4/z6/z8/z10/reg.pt mapped by gallery_merged_cells.npz. Feed order is ckpt['names']; standardize with (X - mu) / sd before scoring.

Eval protocol (numbers in README)

Street arm: OSV-5M 4.89M gallery; 210,122-image sequence-stratified test split. Retrieval: query tower embedding -> proj_head_* -> 512-d, top-K against street_gallery/proj_head_gallery_512_* (same tower space), rerank with the SR/ThrNet head on the 23-feature stack, threshold at 1/25/200/750/2500 km.

Photo arm: unified 14.5M gallery (photo_gallery/emb_*.npy, 512-d fp16); im2gps3k (2997) / yfcc4k (4536) queries; blocklisted honest protocol (184 exact-id leaks removed); 5-fold CV for the headline numbers.