52905e73f6
Stage 2 of the plan is "identity as retrieval", so the first thing measured was whether retrieval can even work here. It cannot, on its own: an off-the-shelf DINOv2 embedding does group identical products (two orange juices, two apple juices, two almond drinks) but the fine-grained cases collapse - 32 cartons into one cluster, and nine different bottled drinks grouped by *container shape*. That matches the literature (RP2K, RPC all fine-tune; nobody ships off-the-shelf features for SKU identity). Meanwhile the shelf has been carrying the answer. Every product has a tag printed under it, and the tag spells the product's name out in clean text, once per product - the store's own label. This adds that as an identity channel: - a tag with a readable name is matched against the catalog by name similarity; - facings are assigned to the *nearest tag along their row* (the tags mark where products sit), not by geometric grouping, which was measured to merge different products into a 945 px "product" that then swallowed the wrong tag; - rows are clustered first, because the same column on six shelves would otherwise become one product. Then the demo catalog itself had to be fixed: it listed another store's products (senbei, Tropicana) while the shelf holds mayonnaise, so nothing could match. The catalog for this shelf was read off the tags - 12 products with names and prices - and the products were enrolled as packaging templates by cropping the facings the tags had just named. Measured through the real still path on add/image6.jpg (111 detections): tag named the product 28 facings (all 28 verified correct by eye) packaging named the product 74 facings both spoke 23 facings, agreeing 16 (69 %) facings with any identity 79 So packaging recognition makes a ~30 % error rate on the facings the tag can check, and every disagreement carried a low packaging score. The tag therefore wins a conflict, and the conflict is recorded (`identity_conflict`) instead of hidden - a product whose tag says one thing and whose packaging says another is a real finding, not noise to suppress. 144 tests pass; test_tag_identity.py adds 6. Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
13 lines
1.2 KiB
CSV
13 lines
1.2 KiB
CSV
SKU_ID,product_name,receipt_date,expiration_date,number_in,number_out,number_on,image_link,price
|
|
sku_l01,キューピー エッグケア 305g,2026-09-18,2026-12-20,36,12,24,'',349
|
|
sku_l02,キューピー アマニ油マヨネーズ 200g,2026-09-18,2026-12-20,24,9,15,'',329
|
|
sku_l03,キューピー からしマヨネーズ 200g,2026-09-18,2026-12-20,30,14,16,'',219
|
|
sku_l04,やまや めんたいマヨネーズタイプ 200g,2026-09-18,2026-11-30,18,6,12,'',399
|
|
sku_l05,キューピー ゼロ ノンコレステロール 310g,2026-09-20,2026-12-25,24,11,13,'',269
|
|
sku_l06,キューピー フィッテ 210g,2026-09-20,2026-12-25,18,7,11,'',379
|
|
sku_l07,丸和油脂 ラッキーマヨネーズ 200g,2026-09-20,2026-12-25,24,10,14,'',199
|
|
sku_l08,スパイシーストーム ペリペリミックス 30g,2026-09-15,2027-03-15,12,4,8,'',699
|
|
sku_l09,ピュアセレクト コクうま カロリーカット65% 360g,2026-09-22,2026-12-28,30,13,17,'',209
|
|
sku_l10,ピュアセレクト マヨネーズ 450g,2026-09-22,2026-12-28,36,15,21,'',222
|
|
sku_l11,キューピー ハーフマヨネーズ 450g,2026-09-19,2026-12-22,48,22,26,'',269
|
|
sku_l12,焼肉のたれ 中辛,2026-09-10,2027-02-10,24,8,16,'',299
|