CodaLab -

> Registration for the use of Pre-trained Models (NLP / Vision)

Dear all,

For a fair competition between all participants. You're required to register for the use of pre-trained models (NLP / Vision).
Please specifically list the pre-trained models you used. A general reference to a whole repository is not acceptable (e.g., a link to TorchVision).

Please simply reply in this forum to register for pre-trained models. For example:

===
01. Pre-trained word embeddings: <link-to-public-repo>
==

Regards,
/The MC-OCR Team

Posted by: mc-ocr-organizers @ Jan. 10, 2021, 10:26 p.m.

Dear all,

As you might get the notification via email that, registration deadline for pre-trained models is on Jan 18, 2021.
By the Jan 18th, 2021, if you don't register for the pre-trained models, your results might not be accepted for the final prizes.

Regards,

Posted by: mc-ocr-organizers @ Jan. 11, 2021, 11:10 p.m.

Dear BTC,

I am currently exploring many pre-trained models that are still being tested but will use the pre-trained ones after tesseract, EAST, HED,CRAFT.
1.Tesseract<https://github.com/tesseract-ocr/tessdata>
2.EAST<https://github.com/argman/EAST>
3.CRAFT<https://github.com/clovaai/CRAFT-pytorch>
4.HED<https://github.com/buntyke/pytorch-hed>

Thanks

Kha

Posted by: VanHoangKha @ Jan. 13, 2021, 6:25 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- PhoBERT: https://github.com/VinAIResearch/PhoBERT

Thanks
Hoang, SUN-AI team

Posted by: phanhoang @ Jan. 13, 2021, 9:53 a.m.

Dear BTC,

I am currently exploring many pre-trained models that are still being tested but will use the pre-trained ones after tesseract, EAST, HED,CRAFT.
1.yolov5<https://github.com/ultralytics/yolov5>
2.deeptext<https://github.com/clovaai/deep-text-recognition-benchmark>
3.CRAFT<https://github.com/clovaai/CRAFT-pytorch>

Thanks

Tung-nguyen, vision-ocr

Posted by: Tung-nguyen @ Jan. 15, 2021, 1:03 a.m.

List of pre-trained models my team might use:

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- PhoBert: https://github.com/VinAIResearch/PhoBERT
- VnCoreNLP: https://github.com/vncorenlp/VnCoreNLP
- Detectron: https://github.com/facebookresearch/Detectron

Posted by: emcadoicobietcodelagi @ Jan. 16, 2021, 11:26 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

Thanks
Hieu, BK_OCR team

Posted by: BK_OCR @ Jan. 16, 2021, 3:55 p.m.

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- PhoBERT: https://github.com/VinAIResearch/PhoBERT
- VnCoreNLP: https://github.com/vncorenlp/VnCoreNLP
- Detectron2: https://github.com/facebookresearch/detectron2
- EfficientNet: https://github.com/lukemelas/EfficientNet-PyTorch
- https://github.com/eragonruan/text-detection-ctpn
- https://github.com/NathanUA/U-2-Net
- https://underthesea.readthedocs.io/en/latest/readme.html

Posted by: phamnhattruong.skyo @ Jan. 17, 2021, 3:28 a.m.

- PaddleOCR: https://github.com/PaddlePaddle/PaddleOCR
- LayoutLM: https://github.com/microsoft/unilm/tree/master/layoutlm
- VietOCR: https://github.com/pbcquoc/vietocr

Posted by: SDSV_AICR @ Jan. 18, 2021, 1:09 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- Yolov5: https://github.com/ultralytics/yolov5
Thanks

Posted by: Nigma.Miracle @ Jan. 18, 2021, 2:22 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models that our team currently used:

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- PaddleOCR: https://github.com/PaddlePaddle/PaddleOCR

Posted by: tiendv @ Jan. 18, 2021, 4:12 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

- Tesseract: https://github.com/madmaze/pytesseract
- PaddleOCR: https://github.com/PaddlePaddle/PaddleOCR
- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- Yolov5: https://github.com/ultralytics/yolov5
- Detectron2: https://github.com/facebookresearch/detectron2

Thanks
Duc, VCC-Analytics team

Posted by: ducnt97 @ Jan. 18, 2021, 8:19 a.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- VietOCR: https://github.com/pbcquoc/vietocr
- Yolov5: https://github.com/ultralytics/yolov5
- LayoutLM: https://github.com/microsoft/unilm/tree/master/layoutlm
- PANnet: https://github.com/WenmuZhou/PAN.pytorch
Thanks

Posted by: litmitc @ Jan. 18, 2021, 11:29 a.m.

Dear MC-OCR organizer committees,

Here is the updated list of pre-trained models in which my team currently used:

Thanks
Hoai, SUN-AI team

Posted by: hoainv @ Jan. 18, 2021, 2:21 p.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

- CRAFT for text detection: https://github.com/clovaai/CRAFT-pytorch
- VietOCR for OCR: https://github.com/pbcquoc/vietocr
- Tesseract for OCR: https://github.com/tesseract-ocr/tessdata
- ETNLP (sent2vec model on 20GB text) for textual embeddings: https://github.com/vietnlp/etnlp
- EfficientNet for visual features: https://github.com/lukemelas/EfficientNet-PyTorch

Bests,

Posted by: thanhduc @ Jan. 18, 2021, 3:50 p.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:

Posted by: haituuuuu @ Jan. 18, 2021, 4:06 p.m.

Dear MC-OCR organizer committees,

Here is the list of pre-trained models in which my team currently used:
- PaddleOCR: https://github.com/PaddlePaddle/PaddleOCR
- LayoutLM: https://github.com/microsoft/unilm/tree/master/layoutlm
- VietOCR: https://github.com/pbcquoc/vietocr
- CRAFT: https://github.com/clovaai/CRAFT-pytorch
- Detectron2: https://github.com/facebookresearch/detectron2
- HuggingFace-BERT: https://huggingface.co/transformers/model_doc/bert.html#
- BPEMB: https://nlp.h-its.org/bpemb/
- EasyOCR: https://github.com/JaidedAI/EasyOCR

Regards

Posted by: pushpendra @ Jan. 19, 2021, 10:05 a.m.

Hi organizers,

Sorry for the late
Our team use pretrained models below:

- mmdetection (Faster R-CNN)
- VietOCR

Posted by: E10.8 @ Jan. 20, 2021, 6:55 p.m.

Dear organizer,
Sorry for late update, our team also used pretrained VGG16.

Posted by: BK_OCR @ Jan. 23, 2021, 3:08 a.m.

Post in this thread

Forums

Mobile-Captured Image Document Recognition for Vietnamese Receipts (MC-OCR) Forum

> Registration for the use of Pre-trained Models (NLP / Vision)