chipoint-2: street-view + general-photo arms, organized

.gitattributes

@@ -1,16 +1,9 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
 *.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
 *.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
 *.npy filter=lfs diff=lfs merge=lfs -text
 *.npz filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
@@ -21,15 +14,14 @@
 *.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
 *.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
 *.tar filter=lfs diff=lfs merge=lfs -text
+*.tf filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
 *.wasm filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.txt filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,332 @@
+---
+
+license: mit
+
+tags:
+
+- image-geolocation
+
+- street-view
+
+- osv-5m
+
+pipeline_tag: image-feature-extraction
+
+datasets:
+
+- osv5m/osv5m
+
+- Jia-py/MP16-Pro
+
+model-index:
+
+- name: Chipoint v2
+
+  results:
+
+  - task:
+
+      type: image-feature-extraction
+
+      name: Image Geolocation
+
+    dataset:
+
+      type: osv5m
+
+      name: OSV-5M
+
+    metrics:
+
+    - name: Acc@1km
+
+      type: accuracy
+
+      value: 2.66
+
+    - name: Acc@25km
+
+      type: accuracy
+
+      value: 44.83
+
+    - name: Acc@200km
+
+      type: accuracy
+
+      value: 81.10
+
+    - name: Acc@750km
+
+      type: accuracy
+
+      value: 92.11
+
+    - name: Acc@2500km
+
+      type: accuracy
+
+      value: 96.82
+
+    - name: GeoScore
+
+      type: geoscore
+
+      value: 4485
+
+    - name: Median error (km)
+
+      type: distance-error
+
+      value: 32
+
+    - name: Mean error (km)
+
+      type: distance-error
+
+      value: 383
+
+  - task:
+
+      type: image-feature-extraction
+
+      name: Image Geolocation (cross-dataset generalization)
+
+    dataset:
+
+      type: im2gps3k
+
+      name: im2gps3k (street-view rerank)
+
+    metrics:
+
+    - name: Acc@25km
+
+      type: accuracy
+
+      value: 11.68
+
+    - name: GeoScore
+
+      type: geoscore
+
+      value: 2018.5
+
+- name: Chipoint v2 (general-photo arm)
+
+  results:
+
+  - task:
+
+      type: image-feature-extraction
+
+      name: Image Geolocation
+
+    dataset:
+
+      type: im2gps3k
+
+      name: im2gps3k (photo)
+
+    metrics:
+
+    - name: Acc@1km
+
+      type: accuracy
+
+      value: 16.5
+
+    - name: Acc@25km
+
+      type: accuracy
+
+      value: 41.2
+
+    - name: Acc@200km
+
+      type: accuracy
+
+      value: 54.1
+
+    - name: Acc@750km
+
+      type: accuracy
+
+      value: 70.7
+
+    - name: Acc@2500km
+
+      type: accuracy
+
+      value: 84.8
+
+  - task:
+
+      type: image-feature-extraction
+
+      name: Image Geolocation
+
+    dataset:
+
+      type: yfcc4k
+
+      name: YFCC4k (photo)
+
+    metrics:
+
+    - name: Acc@1km
+
+      type: accuracy
+
+      value: 19.1
+
+    - name: Acc@25km
+
+      type: accuracy
+
+      value: 34.5
+
+    - name: Acc@200km
+
+      type: accuracy
+
+      value: 44.2
+
+    - name: Acc@750km
+
+      type: accuracy
+
+      value: 60.2
+
+    - name: Acc@2500km
+
+      type: accuracy
+
+      value: 76.0
+
+---
+
+
+
+<p align="center"><img src="wordmark.svg" alt="chiikabu labs" width="340"></p>
+
+
+
+<p align="center"><a href="https://chiikabu.cc">chiikabu.cc</a> &nbsp;·&nbsp; <a href="https://chiikabu.cc/discord">discord</a> &nbsp;·&nbsp; <a href="https://x.com/Chiikabu_Labs">twitter / x</a></p>
+
+
+
+Chipoint v2 is a SOTA geolocation model that is open-weight and beats the
+
+published frontier closed-weight model (Pinpoint, June 2026), its predecessor (Chipoint v1, July 2026),and all models
+
+on the OSV-5M street-view benchmark. This repo hosts both arms:
+
+
+
+- **street-view arm** — the 4.89M-frame OSV-5M gallery index, the
+
+  projection heads, and the 23-feature reranker pipeline
+
+  (`streetview_k512_joint.pt`, K512, per-threshold confidence head)
+
+- **general-photo arm** — the unified 14.5M photo gallery (MP14 10.9M +
+
+  MP16-Pro 4.1M) and the 23-feature rerankers
+
+  (`photoarm_k768_thr.pt` / `photoarm_k768_sr.pt` / `photoarm_k512_thr.pt`)
+
+- `router_osv_flickr.pt` routes queries between the two arms
+
+
+
+## Street-view arm
+
+
+
+![chipoint_v2_mean_error](images/chipoint_v2_mean_error.png)
+
+![chipoint_v2_geoscore](images/chipoint_v2_geoscore.png)
+
+
+
+### OSV-5M benchmark
+
+
+
+OSV-5M test set (210,122 images). Acc@Xkm is the share of predictions
+
+within X km. GeoScore is `5000·exp(-d/1492.7)`, higher is better.
+
+
+
+| Model | @25km | @200km | @750km | @2500km | GeoScore | Mean err |
+
+|---|---|---|---|---|---|---|
+
+| OSV-5M Baseline | – | – | – | – | 3361 | 1814 km |
+
+| GeoCLIP | 21.5 | 52.1 | 72.1 | – | – | – |
+
+| GRE | 9.7 | 35.6 | 72.5 | 91.1 | – | 1192 km |
+
+| LocDiff | 11.0 | 46.3 | 77.0 | 88.2 | – | – |
+
+| RFM (S2S) | – | – | – | – | 3767 | 1069 km |
+
+| HierLoc | – | – | – | – | 3963 | 861 km |
+
+| Pinpoint (retrieval only) | 32.1 | 65.6 | 82.8 | 92.8 | 4035 | 784 km |
+
+| Pinpoint (full) | 35.6 | 67.5 | 83.7 | 93.2 | 4114 | 743 km |
+
+| Chipoint v1 | 37.63 | 70.56 | 84.94 | 93.36 | 4174 | 716 km |
+
+| **Chipoint v2 · street (full pipeline)** | **44.8** | **81.1** | **92.1** | **96.8** | **4485** | **383 km** |
+
+
+
+Chipoint v2's median error is **32 km** (v1: 51 km). The full-pipeline
+
+line is retrieval (K=64 pool) + the K512 23-feature reranker with a
+
+multiscale blend on the full 210,122-image test split. For reference, raw
+
+retrieval without reranking scores 34.89 / 61.96 / 74.60 / 85.34,
+
+GeoScore 3742, mean 1427 km.
+
+
+
+## General-photo arm
+
+
+
+![chipoint_v2_im2gps3k_acc1km](images/chipoint_v2_im2gps3k_acc1km.png)
+
+![chipoint_v2_yfcc4k_acc1km](images/chipoint_v2_yfcc4k_acc1km.png)
+
+
+
+
+
+# Running the model
+
+
+
+Instructions on how to run Chipoint v2 locally are in the RUN_GUIDE.md
+
+file. Quick check:
+
+
+
+```bash
+
+python run_chipoint.py streetview_k512_joint.pt
+
+python run_chipoint.py photoarm_k768_thr.pt
+
+```
+
+
+

RUN_GUIDE.md

@@ -0,0 +1,44 @@
+# 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
+```bash
+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.

features/gallery_merged_cells.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dfa61b5ba9d7086d3db19c3347cb78987f69b892cda7583f91fc3b65c313465
+size 698012606

photo_gallery/emb_0000.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb8804013eb2384ce5bab97f4277400343a8821c60e71f5f7b0e439f3ea2a8d1
+size 4096000128

photo_gallery/emb_0001.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2790668928699549702be892d08cff6c33caf4d2c634d888e20ae0360cdaead
+size 4096000128

photo_gallery/emb_0002.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a51e73caef1114f6a1f0a9c073dee30578e5cf59a2bc1cda66bc6626be9a3e06
+size 4096000128

photo_gallery/emb_0003.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56950a8f56f81b04066a5daf6f9f1bcf5960590f9746f971f86a9e2194d97d9e
+size 2602903680

photo_gallery/ids_0000.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61530892aabfaffd95ef3dfc8a99aa6d0010099bdded896831769228d675eebd
+size 43062052

photo_gallery/ids_0001.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8aee5d44d56204db6731a881eaa36b20aadef044ce34b569d77e8d046fa9bc0d
+size 43999999

photo_gallery/ids_0002.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e151132d6bca41523ffa06135c107cf5251c571068e06e888da85e7460a3e10
+size 54533694

photo_gallery/ids_0003.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:827ec4ef79b65a835ea327671cfd1104ae7e5d3ef6ec3ec8d238daa262a1a86b
+size 53096695

photo_gallery/ll_0000.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eaebf80204bf1f698e18b44bf19d86afedc8f936fc334e756b9b8090515a431
+size 32000128

photo_gallery/ll_0001.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:670d65bbb3c467a41e5ead391085377408086599fc72526450036e64dd7a5c77
+size 32000128

photo_gallery/ll_0002.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98573b23f6353114ae3f2798e2cd947e38de87df93fdb633deb7c8f773dcca0a
+size 32000128

photo_gallery/ll_0003.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70655380bfba7dd9dc6939685dee34cf90cbdfdee1a28f29e34a8cf631f9fdfe
+size 20335312

photo_gallery/manifest.json

@@ -0,0 +1,17 @@
+{
+ "emb_0000.npy": 2048000000,
+ "emb_0001.npy": 2048000000,
+ "emb_0002.npy": 2048000000,
+ "emb_0003.npy": 1301451776,
+ "total_unique": 14541898,
+ "sources": [
+  "mp14_fetch",
+  "mp14_archive",
+  "mp16"
+ ],
+ "source_rows": [
+  10382693,
+  541642,
+  4110903
+ ]
+}

run_chipoint.py

@@ -0,0 +1,74 @@
+"""Load chipoint-2 reranker arms. Run: python run_chipoint.py <ckpt.pt>"""
+import sys
+import torch
+import torch.nn as nn
+
+
+class Blk(nn.Module):
+    def __init__(s, d, h, p=0.1):
+        super().__init__(); s.h = h; s.dk = d // h
+        s.n1 = nn.LayerNorm(d); s.n2 = nn.LayerNorm(d)
+        s.qkv = nn.Linear(d, 3 * d); s.proj = nn.Linear(d, d); s.dr = nn.Dropout(p)
+        s.ff = nn.Sequential(nn.Linear(d, 4 * d), nn.GELU(), nn.Dropout(p), nn.Linear(4 * d, d))
+
+    def forward(s, x):
+        B, K, D = x.shape
+        q, k, v = s.qkv(s.n1(x)).chunk(3, -1)
+        q, k, v = (t.view(B, K, s.h, s.dk).transpose(1, 2) for t in (q, k, v))
+        a = torch.nn.functional.scaled_dot_product_attention(q, k, v)
+        x = x + s.dr(s.proj(a.transpose(1, 2).reshape(B, K, D)))
+        return x + s.ff(s.n2(x))
+
+
+class SR(nn.Module):
+    def __init__(s, nf, d=256, h=4, l=3, p=0.1):
+        super().__init__()
+        s.inp = nn.Sequential(nn.Linear(nf, d), nn.GELU(), nn.LayerNorm(d))
+        s.b = nn.ModuleList([Blk(d, h, p) for _ in range(l)])
+        s.n = nn.LayerNorm(d); s.o = nn.Linear(d, 1)
+
+    def forward(s, x):
+        h = s.inp(x)
+        for b in s.b:
+            h = b(h)
+        return s.o(s.n(h)).squeeze(-1)
+
+
+class ThrNet(nn.Module):
+    def __init__(s, nf, d=256, h=4, l=3, p=0.1, n_thr=5, per_bin=False):
+        super().__init__()
+        s.inp = nn.Sequential(nn.Linear(nf, d), nn.GELU(), nn.LayerNorm(d))
+        s.b = nn.ModuleList([Blk(d, h, p) for _ in range(l)])
+        if per_bin:
+            s.ns = nn.ModuleList([nn.LayerNorm(d) for _ in range(n_thr)])
+            s.os = nn.ModuleList([nn.Linear(d, 1) for _ in range(n_thr)])
+        else:
+            s.n = nn.LayerNorm(d); s.o = nn.Linear(d, n_thr)
+
+    def forward(s, x):
+        h = s.inp(x)
+        for b in s.b:
+            h = b(h)
+        if hasattr(s, 'ns'):
+            return torch.cat([o(n(h)) for n, o in zip(s.ns, s.os)], -1)
+        return s.o(s.n(h))
+
+
+def main():
+    ck = torch.load(sys.argv[1], map_location='cpu', weights_only=False)
+    nf = ck['nf']
+    net = SR(nf)
+    net.load_state_dict(ck['state'])
+    net.eval()
+    print('nf', nf, '| names', len(ck['names']), '| mu/sd', tuple(ck['mu'].shape))
+    if 'thr_state' in ck:
+        per_bin = any(k.startswith('ns.') for k in ck['thr_state'])
+        thr = ThrNet(nf, per_bin=per_bin)
+        thr.load_state_dict(ck['thr_state'])
+        thr.eval()
+        print('thr_km', ck['thr_km'], '| per_bin', per_bin)
+    print('OK — feed (Nq,K,nf) features standardized by ck[mu]/ck[sd]')
+
+
+if __name__ == '__main__':
+    main()

street_gallery/gallery_all_C_so400m256.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57546309fb3fdc7beaaf8ede63190aef10199da0ca2726a528c3e09f65326b11
+size 39157600

street_gallery/proj_head_gallery_512_dinov3.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3527fccb1f1ad6961254105318ad61fdf60a5b935e6e23ecd9e82d658e784552
+size 5012156544

street_gallery/proj_head_gallery_512_gopt.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b46f74e298bda45af3485c727c46e771d67356e2b9afe80fdcb56b3bb423dfd
+size 5012156544

street_gallery/proj_head_gallery_512_so400m256.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30c7af4fb7c7afce39bd0279e15a33c1546d96878f56bb87c5350e8dd48e6890
+size 5012156544

weights/ghead_merged_reg.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd3cc14efb34fd5bd04bfb55bad70740ed353c29c05381c9d4b8e1b0ad49fe06
+size 2129483

weights/ghead_merged_z10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:081524b93a69205fdde5851dfa0cc5b2e521e84bc34bca6e171cec044c0cb0d3
+size 209711307

weights/ghead_merged_z4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ceb1bbdfb320ed8e423c02c78bf648d2bd7fb6132f79f5bb990427f9c56eeb95
+size 3166781

weights/ghead_merged_z6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3458bb052ee4853e04e3d7b67f1c46b57796b4fcad576c9ea126c273253b798d
+size 18906685

weights/ghead_merged_z8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:119c69def5330b516ccad9bcd89fe5836e723eaacd883e5ec856f31e1e336ee5
+size 270794237

weights/photoarm_k512_thr.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:795ee161e2fe6a11875e64af9879e20a07a6f53c3fc38950a618f2cd7c56db11
+size 19047925

weights/photoarm_k768_sr.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ab63b83bc9c906e57cb168dc29ec0ec93138df3366593eef02473e7cc524ae8
+size 9523129

weights/photoarm_k768_thr.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a8266ab2c8580f98e19d301484f433e82df033a7ab066a7ac9c07c3bfd43ca3
+size 19061275

weights/proj_head_dinov3.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e88528e1f8d1a06669a36764db69d1dff6c14c0a62eb9a37ce5c76c282f62daa
+size 139420540

weights/proj_head_gopt.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36a399124ead993f00444b530a2e9b5526868fe2655d16cad640a6ab6ad8c814
+size 110797670

weights/proj_head_so400m256.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41dfc4f87bfa19edcc8664123869bb782ca0dcfffbe7153129b2444efe48d1b4
+size 109224861

weights/router_osv_flickr.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea57124d833076d7562f1d69179f9ca346b680c98bce82b8ac4a9f397c0dcbf7
+size 271617

weights/streetview_k512_joint.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04952d4d025e3a4ea081871af5066e2345821aa9661ca636deb1bcb033c3930f
+size 19047925