tngo/product-recognition-demo
1
0
Fork 0
forked from vnguyen/product-recognition-demo
Code Pull requests Activity

docs: name the local lane in the evaluators' own lane vocabulary

Browse source
This commit is contained in:
tngo 2026-09-28 22:34:06 +09:00
parent 4aa1b35e39
commit 3c0df81629
2 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
recognization/code/eval_identity.py
View file

@ -3,9 +3,9 @@
The original matcher is a candidate final architecture, not just a baseline. The original matcher is a candidate final architecture, not just a baseline.
Every lane is scored the same way, on the same crops, at the same resolutions: Every lane is scored the same way, on the same crops, at the same resolutions:
local RootSIFT pyramid + OCR text lane (the shipped default)
matcher full-catalog SIFT/ORB/BRISK + RANSAC (the original demo) matcher full-catalog SIFT/ORB/BRISK + RANSAC (the original demo)
gallery DINOv2 embedding, raw or with the trained head gallery DINOv2 embedding, raw or with the trained head
hybrid-verify DINO shortlist, then the matcher re-ranks inside it
Reported per lane, as four separate ceilings rather than one conflated number: Reported per lane, as four separate ceilings rather than one conflated number:
@ -56,7 +56,7 @@ UNKNOWN = "unknown"
# Two independent label passes disagreed here, so the row is not ground truth. # Two independent label passes disagreed here, so the row is not ground truth.
# Scoring it would be scoring a coin flip. # Scoring it would be scoring a coin flip.
EXCLUDE = "exclude" EXCLUDE = "exclude"
LANES = ("matcher", "gallery", "hybrid-verify") LANES = ("local", "matcher", "gallery", "hybrid-verify")
# A single SKU above this share of accepts is the failure the last audit found: # A single SKU above this share of accepts is the failure the last audit found:
# every dark package collapsing onto one enrolled product. # every dark package collapsing onto one enrolled product.
HUB_SHARE = 0.30 HUB_SHARE = 0.30

2
recognization/code/eval_live.py
View file

@ -11,7 +11,7 @@ store was enrolled; scoring them is training-set accuracy.
python code/eval_live.py RECORDING_DIR [--stock-dir DIR] [--labels PATH] python code/eval_live.py RECORDING_DIR [--stock-dir DIR] [--labels PATH]
[--json PATH] [--iou 0.5] [--json PATH] [--iou 0.5]
[--replay] [--identity matcher|gallery] [--device cpu|mps|cuda] [--replay] [--identity matcher|local|gallery] [--device cpu|mps|cuda]
Prints a human count table and writes the same report as JSON (default Prints a human count table and writes the same report as JSON (default
<recording>/eval_live.json). Missing input exits 2 with {"error": ...} on <recording>/eval_live.json). Missing input exits 2 with {"error": ...} on