Co-compiled Edge TPU sets root-plate-bird, face-plate-bird, plate-bird (CamStack D722)

edgetpuSets/face-plate-bird/NOTICE-open-lpr.txt

@@ -0,0 +1,44 @@
+Source: https://huggingface.co/faisalthaheem/open-lpr-models/resolve/aed0df35ec2891c5a90437041ebeadb592cc516b/NOTICE
+Fetched: 2026-10-06 (verbatim below the line; nothing edited)
+Applies to: plateDetection/yolox-tiny-plate (open-lpr YOLOX-tiny plate detector, Apache-2.0; licence text: Apache-2.0.txt). Apache-2.0 section 4(d) requires this NOTICE to accompany redistribution.
+--------------------------------------------------------------------------------
+open-lpr models
+Copyright the open-lpr contributors
+
+This product includes software developed by third parties. The artifacts in
+this repository are NOT all ours; per-artifact provenance is recorded in
+manifest.json and summarised below.
+
+--------------------------------------------------------------------------------
+plate_yolox_tiny_640.onnx  and  best_ckpt.pth
+--------------------------------------------------------------------------------
+Produced by this project. Architecture from YOLOX (Megvii-BaseDetection/YOLOX),
+which is licensed Apache-2.0 for both its code and its same-repository release
+weights. Pretrained from COCO (yolox_tiny.pth), then fine-tuned on an annotated
+licence-plate corpus plus community-contributed images pseudo-labelled by a
+teacher model.
+
+The training imagery is not published. It consists of user-contributed
+photographs containing real licence plates, faces and vehicles, uploaded without
+consent obtained. Publishing it would create a searchable index of real vehicle
+registration numbers.
+
+Ultralytics YOLO weights are licensed AGPL-3.0 and are deliberately NOT used,
+redistributed, or derived from anywhere in this project.
+
+--------------------------------------------------------------------------------
+plate_ocr_ppocrv5_mobile.onnx  and  plate_ocr_dict.json
+--------------------------------------------------------------------------------
+NOT produced by this project. These are conversions of an upstream recognition
+model from PaddleOCR (https://github.com/PaddlePaddle/PaddleOCR), which is
+licensed under Apache-2.0.
+
+The specific upstream release was never recorded and cannot be recovered from
+the artifacts: the ONNX graph carries no producer, producer_version or
+doc_string, and no PaddlePaddle installation, wheel or shell history survives on
+the build host. The Apache-2.0 licence is therefore asserted at project level
+and is NOT pinned to a verified release.
+
+These two files are a MATCHED PAIR and must be used together. Decoding assumes
+len(dictionary) + 2 output classes; substituting either without the other
+produces silently incorrect text rather than an error.

edgetpuSets/face-plate-bird/README.md

@@ -0,0 +1,30 @@
+---
+license: other
+tags:
+  - edge-tpu
+  - coral
+  - camstack
+---
+
+# Co-compiled Edge TPU set `face-plate-bird` — CamStack
+
+Files for ONE Coral, compiled together by `edgetpu_compiler` 14.1 so they share
+the Edge TPU parameter cache instead of evicting each other (CamStack D722). A
+CamStack Coral pool loads this set when the models placed on it are covered by
+it; load the files of one set on one Coral, never mixed with other sets or with
+the separately compiled builds. Built by `scripts/build-edgetpu-sets.py`.
+
+| Member (compile order = cache priority) | Catalog model | On chip | Streamed per invoke |
+| --- | --- | --- | --- |
+| `face-plate-bird__camstack-yunet-2023mar_int8_edgetpu.tflite` | `yunet-2023mar` | 0.11 MiB | 0.00 MiB |
+| `face-plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite` | `yolox-tiny-plate` | 3.23 MiB | 2.27 MiB |
+| `face-plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite` | `bird-classifier` | 0.00 MiB | 4.54 MiB |
+
+Each member is the same integer graph as its catalog model's own `tflite`
+build; only the compilation differs.
+
+## Licences
+
+- YuNet 2023mar, weights MIT (Copyright (c) 2020 Shiqi Yu); training data WIDER FACE (CC BY-NC-ND images)
+- YOLOX-tiny plate detector by the open-lpr project, Apache-2.0 (NOTICE in this folder; training data terms UNKNOWN)
+- Google AIY Vision Birds V1, Apache-2.0

edgetpuSets/face-plate-bird/face-plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e004a9db021139bc353cf03d7cbad3130b10c63be2f387d4eea84dc6cffb7381
+size 6882320

edgetpuSets/face-plate-bird/face-plate-bird__camstack-yunet-2023mar_int8_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf56290776ab506d9c922fe21c69cd483060b361b71dfe21028ad47cc04d7274
+size 714256

edgetpuSets/face-plate-bird/face-plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d2c8c78cdc37128599e43077c1761bb27f2269b66e7ebf69c74aa4ddc1c4f59
+size 4923968

edgetpuSets/plate-bird/NOTICE-open-lpr.txt

@@ -0,0 +1,44 @@
+Source: https://huggingface.co/faisalthaheem/open-lpr-models/resolve/aed0df35ec2891c5a90437041ebeadb592cc516b/NOTICE
+Fetched: 2026-10-06 (verbatim below the line; nothing edited)
+Applies to: plateDetection/yolox-tiny-plate (open-lpr YOLOX-tiny plate detector, Apache-2.0; licence text: Apache-2.0.txt). Apache-2.0 section 4(d) requires this NOTICE to accompany redistribution.
+--------------------------------------------------------------------------------
+open-lpr models
+Copyright the open-lpr contributors
+
+This product includes software developed by third parties. The artifacts in
+this repository are NOT all ours; per-artifact provenance is recorded in
+manifest.json and summarised below.
+
+--------------------------------------------------------------------------------
+plate_yolox_tiny_640.onnx  and  best_ckpt.pth
+--------------------------------------------------------------------------------
+Produced by this project. Architecture from YOLOX (Megvii-BaseDetection/YOLOX),
+which is licensed Apache-2.0 for both its code and its same-repository release
+weights. Pretrained from COCO (yolox_tiny.pth), then fine-tuned on an annotated
+licence-plate corpus plus community-contributed images pseudo-labelled by a
+teacher model.
+
+The training imagery is not published. It consists of user-contributed
+photographs containing real licence plates, faces and vehicles, uploaded without
+consent obtained. Publishing it would create a searchable index of real vehicle
+registration numbers.
+
+Ultralytics YOLO weights are licensed AGPL-3.0 and are deliberately NOT used,
+redistributed, or derived from anywhere in this project.
+
+--------------------------------------------------------------------------------
+plate_ocr_ppocrv5_mobile.onnx  and  plate_ocr_dict.json
+--------------------------------------------------------------------------------
+NOT produced by this project. These are conversions of an upstream recognition
+model from PaddleOCR (https://github.com/PaddlePaddle/PaddleOCR), which is
+licensed under Apache-2.0.
+
+The specific upstream release was never recorded and cannot be recovered from
+the artifacts: the ONNX graph carries no producer, producer_version or
+doc_string, and no PaddlePaddle installation, wheel or shell history survives on
+the build host. The Apache-2.0 licence is therefore asserted at project level
+and is NOT pinned to a verified release.
+
+These two files are a MATCHED PAIR and must be used together. Decoding assumes
+len(dictionary) + 2 output classes; substituting either without the other
+produces silently incorrect text rather than an error.

edgetpuSets/plate-bird/README.md

@@ -0,0 +1,28 @@
+---
+license: other
+tags:
+  - edge-tpu
+  - coral
+  - camstack
+---
+
+# Co-compiled Edge TPU set `plate-bird` — CamStack
+
+Files for ONE Coral, compiled together by `edgetpu_compiler` 14.1 so they share
+the Edge TPU parameter cache instead of evicting each other (CamStack D722). A
+CamStack Coral pool loads this set when the models placed on it are covered by
+it; load the files of one set on one Coral, never mixed with other sets or with
+the separately compiled builds. Built by `scripts/build-edgetpu-sets.py`.
+
+| Member (compile order = cache priority) | Catalog model | On chip | Streamed per invoke |
+| --- | --- | --- | --- |
+| `plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite` | `yolox-tiny-plate` | 5.55 MiB | 0.00 MiB |
+| `plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite` | `bird-classifier` | 0.92 MiB | 2.80 MiB |
+
+Each member is the same integer graph as its catalog model's own `tflite`
+build; only the compilation differs.
+
+## Licences
+
+- YOLOX-tiny plate detector by the open-lpr project, Apache-2.0 (NOTICE in this folder; training data terms UNKNOWN)
+- Google AIY Vision Birds V1, Apache-2.0

edgetpuSets/plate-bird/plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c60fbc659ea873f1469519701c7ed3c2dc8463e3e756c481e7cd985b364f824e
+size 6865936

edgetpuSets/plate-bird/plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:063c5b7862af84132880c24ae62d18271dd2c0bc2942a6a106f8cad04b1f5143
+size 4080192

edgetpuSets/root-plate-bird/NOTICE-open-lpr.txt

@@ -0,0 +1,44 @@
+Source: https://huggingface.co/faisalthaheem/open-lpr-models/resolve/aed0df35ec2891c5a90437041ebeadb592cc516b/NOTICE
+Fetched: 2026-10-06 (verbatim below the line; nothing edited)
+Applies to: plateDetection/yolox-tiny-plate (open-lpr YOLOX-tiny plate detector, Apache-2.0; licence text: Apache-2.0.txt). Apache-2.0 section 4(d) requires this NOTICE to accompany redistribution.
+--------------------------------------------------------------------------------
+open-lpr models
+Copyright the open-lpr contributors
+
+This product includes software developed by third parties. The artifacts in
+this repository are NOT all ours; per-artifact provenance is recorded in
+manifest.json and summarised below.
+
+--------------------------------------------------------------------------------
+plate_yolox_tiny_640.onnx  and  best_ckpt.pth
+--------------------------------------------------------------------------------
+Produced by this project. Architecture from YOLOX (Megvii-BaseDetection/YOLOX),
+which is licensed Apache-2.0 for both its code and its same-repository release
+weights. Pretrained from COCO (yolox_tiny.pth), then fine-tuned on an annotated
+licence-plate corpus plus community-contributed images pseudo-labelled by a
+teacher model.
+
+The training imagery is not published. It consists of user-contributed
+photographs containing real licence plates, faces and vehicles, uploaded without
+consent obtained. Publishing it would create a searchable index of real vehicle
+registration numbers.
+
+Ultralytics YOLO weights are licensed AGPL-3.0 and are deliberately NOT used,
+redistributed, or derived from anywhere in this project.
+
+--------------------------------------------------------------------------------
+plate_ocr_ppocrv5_mobile.onnx  and  plate_ocr_dict.json
+--------------------------------------------------------------------------------
+NOT produced by this project. These are conversions of an upstream recognition
+model from PaddleOCR (https://github.com/PaddlePaddle/PaddleOCR), which is
+licensed under Apache-2.0.
+
+The specific upstream release was never recorded and cannot be recovered from
+the artifacts: the ONNX graph carries no producer, producer_version or
+doc_string, and no PaddlePaddle installation, wheel or shell history survives on
+the build host. The Apache-2.0 licence is therefore asserted at project level
+and is NOT pinned to a verified release.
+
+These two files are a MATCHED PAIR and must be used together. Decoding assumes
+len(dictionary) + 2 output classes; substituting either without the other
+produces silently incorrect text rather than an error.

edgetpuSets/root-plate-bird/README.md

@@ -0,0 +1,30 @@
+---
+license: other
+tags:
+  - edge-tpu
+  - coral
+  - camstack
+---
+
+# Co-compiled Edge TPU set `root-plate-bird` — CamStack
+
+Files for ONE Coral, compiled together by `edgetpu_compiler` 14.1 so they share
+the Edge TPU parameter cache instead of evicting each other (CamStack D722). A
+CamStack Coral pool loads this set when the models placed on it are covered by
+it; load the files of one set on one Coral, never mixed with other sets or with
+the separately compiled builds. Built by `scripts/build-edgetpu-sets.py`.
+
+| Member (compile order = cache priority) | Catalog model | On chip | Streamed per invoke |
+| --- | --- | --- | --- |
+| `root-plate-bird__ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite` | `ssd-mobilenet-v2-coco-edgetpu` | 6.39 MiB | 0.00 MiB |
+| `root-plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite` | `yolox-tiny-plate` | 0.08 MiB | 5.41 MiB |
+| `root-plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite` | `bird-classifier` | 0.00 MiB | 4.54 MiB |
+
+Each member is the same integer graph as its catalog model's own `tflite`
+build; only the compilation differs.
+
+## Licences
+
+- SSD MobileNet V2 COCO, Google Coral test_data, Apache-2.0 (training data: COCO)
+- YOLOX-tiny plate detector by the open-lpr project, Apache-2.0 (NOTICE in this folder; training data terms UNKNOWN)
+- Google AIY Vision Birds V1, Apache-2.0

edgetpuSets/root-plate-bird/root-plate-bird__camstack-yolox-tiny-plate_int8_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f306d56b7024d256fed5639cec490579302e457a038754fae43fd460d85e1e7a
+size 6906896

edgetpuSets/root-plate-bird/root-plate-bird__mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee003f3672f927ee611698daa080d5fcf05744d60fc71be44bf475afe9b15a4e
+size 4923968

edgetpuSets/root-plate-bird/root-plate-bird__ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d0be166f1c06cccc7a2f81df911c3ee47e0d6588d1c866e43988646ac821e0a
+size 6946352