Arulkumar03/Wheat_HEAD_Detection_Counting_ComputerVision_Model

1# DensePose in Detectron22 3DensePose aims at learning and establishing dense correspondences between image pixels4and 3D object geometry for deformable objects, such as humans or animals.5In this repository, we provide the code to train and evaluate DensePose R-CNN and6various tools to visualize DensePose annotations and results.7 8There are two main paradigms that are used within DensePose project.9 10## [Chart-based Dense Pose Estimation for Humans and Animals](doc/DENSEPOSE_IUV.md)11 12<div align="center">13  <img src="https://dl.fbaipublicfiles.com/densepose/web/densepose_teaser_compressed_25.gif" width="700px" />14</div>15 16For chart-based estimation, 3D object mesh is split into charts and17for each pixel the model estimates chart index `I` and local chart coordinates `(U, V)`.18Please follow the link above to find a [detailed overview](doc/DENSEPOSE_IUV.md#Overview)19of the method, links to trained models along with their performance evaluation in the20[Model Zoo](doc/DENSEPOSE_IUV.md#ModelZoo) and21[references](doc/DENSEPOSE_IUV.md#References) to the corresponding papers.22 23## [Continuous Surface Embeddings for Dense Pose Estimation for Humans and Animals](doc/DENSEPOSE_CSE.md)24 25<div align="center">26  <img src="https://dl.fbaipublicfiles.com/densepose/web/densepose_cse_teaser.png" width="700px" />27</div>28 29To establish continuous surface embeddings, the model simultaneously learns30descriptors for mesh vertices and for image pixels.31The embeddings are put into correspondence, thus the location32of each pixel on the 3D model is derived.33Please follow the link above to find a [detailed overview](doc/DENSEPOSE_CSE.md#Overview)34of the method, links to trained models along with their performance evaluation in the35[Model Zoo](doc/DENSEPOSE_CSE.md#ModelZoo) and36[references](doc/DENSEPOSE_CSE.md#References) to the corresponding papers.37 38# Quick Start39 40See [ Getting Started ](doc/GETTING_STARTED.md)41 42# Model Zoo43 44Please check the dedicated pages45for [chart-based model zoo](doc/DENSEPOSE_IUV.md#ModelZoo)46and for [continuous surface embeddings model zoo](doc/DENSEPOSE_CSE.md#ModelZoo).47 48# What's New49 50* June 2021: [DensePose CSE with Cycle Losses](doc/RELEASE_2021_06.md)51* March 2021: [DensePose CSE (a framework to extend DensePose to various categories using 3D models)52  and DensePose Evolution (a framework to bootstrap DensePose on unlabeled data) released](doc/RELEASE_2021_03.md)53* April 2020: [DensePose Confidence Estimation and Model Zoo Improvements](doc/RELEASE_2020_04.md)54 55# License56 57Detectron2 is released under the [Apache 2.0 license](../../LICENSE)58 59## <a name="CitingDensePose"></a>Citing DensePose60 61If you use DensePose, please refer to the BibTeX entries62for [chart-based models](doc/DENSEPOSE_IUV.md#References)63and for [continuous surface embeddings](doc/DENSEPOSE_CSE.md#References).64 65