Monk - A computer vision toolkit for everyone

Monk Object Detection - A low code wrapper over state-of-the-art deep learning algorithms

<br />

Why use Monk

• Issue: Abudance of algorithms and difficult to find a working code

  • <b> Solution: All your state-of-the-art as well as old algorithms in one place </b>

• Issue: Installaing different deep learning pipelines is an error-prone task

  • <b> Solution: Single line installations with monk </b>

• Issue: Setting up different algorithms for your custom data requires a lot of effort in changing the existing codes

  • <b> Solution: Easily ingest your custom data for training in COCO, VOC, or Yolo formats </b>

• Issue: Difficulty to trace out which hyperparameters to change for tuning the algorithm

  • <b> Solution: Set your hyper-parameters with a common structure for every algorithm </b>

• Issue: Deployment requires knowledge of base libraries and codes

  • <b> Solution: Easily deploy your models using Monk's low code-syntax </b>

• Issue: Looking for hands-on tutorials for computer vision

  • <b> Solution: Use monk's application building tutorial set</b>

<br /> <br />

Create real-world Object Detection applications

<table> <tr> <td>Wheat detection in field</td> <td>Detection in underwater imagery</td> <td>Trash Detection</td> </tr> <tr> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/wheat-detection-demo.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/sea_tutrle_demo.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/trash.gif" width=320 height=240></td> </tr> <tr> <td>Object detection in bad lighting</td> <td>Tiger detection in wild</td> <td>Person detection in infrared imagery</td> </tr> <tr> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/obj-det-in-bad-light.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/tiger.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/ir-person-det.gif" width=320 height=240></td> </tr> </table>

For more such tutorials visit Application Model Zoo

<br /> <br />

Create real-world Image Segmentation applications

<table> <tr> <td>Road Segmentation in satellite imagery</td> <td>Ultrasound nerve segmentation</td> </tr> <tr> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/satellite-road-segmentation.gif" width=640 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/ultrasound-nerve-image-segmentat.gif" width=320 height=240></td> </tr> </table>

For more such tutorials visit Application Model Zoo

<br /> <br />

Other applications

<table> <tr> <td>Face Detection</td> <td>Pose Estimation</td> <td>Activity Recognition</td> </tr> <tr> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/face.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/pose_demo.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/ucf-demo.gif" width=320 height=240></td> </tr> <tr> <td>Object Re-identification</td> <td>Scene Text Localization</td> <td>Object Tracking</td> </tr> <tr> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/coming_soon.jpg" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/text_demo.gif" width=320 height=240></td> <td><img src="https://github.com/abhi-kumar/monk_det_demos/blob/master/coming_soon.jpg" width=320 height=240></td> </tr> </table>

For more such tutorials visit Application Model Zoo

<br /> <br />

Important Elements

• A) Training Engine

  • Train models on custom dataset witjh low code syntax
  • Pretrained examples on variety of datasets
  • Useful to train your own detector

• B) Inference Engine

  • Original pretrained models (from original authors and implementations) for inferencing and analysing
  • Pretrained models on coco, voc, cityscpaes, type datasets
  • Useful to analyse which algorithm works best for you
  • Useful to generate semi-accurate annotations (coco, pascal-voc, yolo formats) on a new dataset

<br /> <br /> <br />

Training Engine Algorithms

- Train models on custom dataset with low code syntax
- Pretrained examples on variety of datasets
- Useful to train your own detector

NOTE - See the licence file mentioned in the pipelines before using them

S.No.	Algorithm Type	Algorithm	Model variations	Installation	Example Notebooks	Code	Credits	Original Usage License	Functional Docs
1	Object Detection	GluonCV Finetune	5	LINK	LINK	LINK	LINK	Apache 2.0	LINK
2	Object Detection	Tensorflow Object Detection 1.0	22	LINK	LINK	LINK	LINK	Apache 2.0	In Development
3	Object Detection	Tensorflow Object Detection 2.0	26	LINK	LINK	LINK	LINK	Apache 2.0	In Development
4	Object Detection	Pytorch Efficient-Det 1	1	LINK	LINK	LINK	LINK	MIT	LINK
5	Object Detection	Pytorch Efficient-Det 2	8	LINK	LINK	LINK	LINK	LGPL 3.0	In Development
6	Object Detection	TorchVision Finetune	1	LINK	LINK	LINK	LINK	BSD-3-Clause	LINK
7	Object Detection	Mx-RCNN	3	LINK	LINK	LINK	LINK	Mixed	LINK
8	Object Detection	Pytorch-Retinanet	5	LINK	LINK	LINK	LINK	Apache 2.0	LINK
9	Object Detection	CornerNet Lite	2	LINK	LINK	LINK	LINK	BSD-3-Clause	LINK
10	Object Detection	YoloV3	7	LINK	LINK	LINK	LINK	GPL 3.0	LINK
11	Object Detection	RFBNet	3	LINK	LINK	LINK	LINK	MIT	LINK
12	Object Detection	Slim-Yolo-V3	1	LINK	LINK	LINK	LINK	License Not Available	In Development
13	Object Detection	Pytorch SSD	3	LINK	LINK	LINK	LINK	MIT	In Development
14	Object Detection	Pytorch-Peleenet	1	LINK	LINK	LINK	LINK	License Not Available	In Development
15	Object Detection	MM-Detection	36	LINK	LINK	LINK	LINK	Apache 2.0	In Development
16	Image Segmentation	Segmentation Models	4	LINK	LINK	LINK	LINK	MIT	In Development
17	Pytorch Retinaface	Face Detection	2	LINK	LINK	LINK	LINK	MIT	In Development
18	Action Recognition	MM-Action2	8	LINK	LINK	LINK	LINK	Apache 2.0	In Development
19	Text Localization	Pytorch-TextSnake	1	LINK	LINK	LINK	LINK	MIT	In Development
20	Image Segmentation	SOLO - V1/V2	14	LINK	LINK	LINK	LINK	Academic non-commercial usage	In Development
21	Image Segmentation	Mask-RCNN (MMDetect)	8	LINK	LINK	LINK	LINK	Apache 2.0	In Development
22	Pose Estimation	GluonCV Pose	11	LINK	LINK	LINK	LINK	Apache 2.0

<br /> <br /> <br />

Inference Engine Algorithms

- Infer already trained models on COCO/VOC/Open-Images on your custom data
- Useful to analyse computation time metrics

S.No.	Algorithm Type	Algorithm	Model Valriations	Model Trained On	Installation	Example Notebook	Code	Credits	Functional Docs
1	Object Detection	GluonCV Finetune	4	COCO	Pascal VOC	LINK	LINK	LINK	LINK
2	Object Detection	Pytorch EfficientDet	8	COCO	LINK	LINK	LINK	LINK	In Development
3	Object Detection	Detecto-RS	2	COCO	LINK	LINK	LINK	LINK	In Development

<br /> <br /> <br />

Aknowledgements

• Contributors' information can be found here: https://github.com/Tessellate-Imaging/Monk_Object_Detection/blob/master/Contributors.md
• Majority of the code is obtained from these pipelines (Monk is a low code wrapper to utilize these pipelines)
  • https://gluon-cv.mxnet.io/build/examples_detection/index.html
  • https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1.md
  • https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2.md
  • https://github.com/signatrix/efficientdet
  • https://github.com/zylo117/Yet-Another-EfficientDet-Pytorch
  • https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html
  • https://github.com/ijkguo/mx-rcnn
  • https://github.com/yhenon/pytorch-retinanet
  • https://github.com/princeton-vl/CornerNet-Lite
  • https://github.com/ultralytics/yolov3
  • https://github.com/ruinmessi/RFBNet
  • https://github.com/PengyiZhang/SlimYOLOv3
  • https://github.com/qfgaohao/pytorch-ssd
  • https://github.com/open-mmlab/mmdetection
  • https://github.com/qubvel/segmentation_models
  • https://github.com/biubug6/Pytorch_Retinaface
  • https://github.com/open-mmlab/mmaction2
  • https://github.com/WXinlong/SOLO
  • https://gluon-cv.mxnet.io/build/examples_pose/dive_deep_simple_pose.html
  • https://github.com/Cartucho/mAP

Author

Tessellate Imaging - https://www.tessellateimaging.com/

Check out Monk AI - (https://github.com/Tessellate-Imaging/monk_v1)

Monk features
    - low-code
    - unified wrapper over major deep learning framework - keras, pytorch, gluoncv
    - syntax invariant wrapper

Enables developers
    - to create, manage and version control deep learning experiments
    - to compare experiments across training metrics
    - to quickly find best hyper-parameters

To contribute to Monk AI or Monk Object Detection repository raise an issue in the git-repo or dm us on linkedin

• Abhishek - https://www.linkedin.com/in/abhishek-kumar-annamraju/
• Akash - https://www.linkedin.com/in/akashdeepsingh01/ <br />

<br /> <br />

Copyright

Copyright 2019 onwards, Tessellate Imaging Private Limited Licensed under the Apache License, Version 2.0 (the "License"); you may not use this project's files except in compliance with the License. A copy of the License is provided in the LICENSE file in this repository.