/cv-problem-based-algorithms

👨‍💻 problem based computer vision algorithms

For computer vision datasets,CNN architecture

cv_algorithms

Typical tasks

Recognition

Determining whether or not the image data contains some specific object, feature, or activity.

Object recognition (also called object classification)

one or several pre-specified or learned objects or object classes can be recognized, usually together with their 2D positions in the image or 3D poses in the scene. Blippar, Google Goggles and LikeThat provide stand-alone programs that illustrate this functionality.

Objects detection

Object detection

After 2014 – Deep Learning Detection period

Most important two-stage object detection algorithms

RCNN (2014)

SPPNet (2014)

Fast RCNN (2015)

Faster RCNN (2015)

Mask-RCNN(2017)

Feature Pyramid Networks/FPN (2017)

D2Det(2020)

G-RCNN (2021)

Sparse R-CNN(2021)

Most important one-stage object detection algorithms

You Only Look Once (YOLO) (2016)

YOLO2 (2017)

Single Shot MultiBox Detector (SSD) (2016)

RetinaNet (2017)

YOLOv3 (2018)

CornerNet (2018)

CenterNet

Trident Net(2019)

YOLOv4 (2020)

EffcientNet (2020)

CentripetalNet (2020)

YOLOR (2021)

Yolo7(2022)

mmdetection

Before 2014 – Traditional Object Detection period

Deformable Part-based Model (DPM) (2008) with the first introduction of bounding box regression

HOG Detector (2006), a popular feature descriptor for object detection in computer vision and image processing

Viola-Jones Detector (2001), the pioneering work that started the development of traditional object detection methods

References

Identification / individual instance of an object is recognized

Content-based image retrieval

References

Pose estimation

Optical character recognition (OCR)

2D code reading

Facial recognition

Shape Recognition Technology (SRT)

Motion analysis

Egomotion

video tracking/ object tracking

[Simple Online And Realtime Tracking] (SORT)(https://github.com/abewley/sort)

DeepSORT

FairMOT

[TransMOT] (https://arxiv.org/pdf/2104.00194v2.pdf)

[ByteTrack] (https://arxiv.org/pdf/2110.06864v2.pdf)

Optical flow

Scene reconstruction

Image restoration

Image Generation

De-noising

Super-Resolution

3D Face Animation

Action Classification in Video

Image segmentation

Edge detection

Image restoration