BlenderProc2 is a procedural pipeline that can render realistic images for the training of neural networks. Our pipeline can be employed in various use cases, including segmentation, depth, normal and pose estimation, and many others. A key feature of our Blender extension is the simple-to-use python API, designed to be easily extendable. Furthermore, many public datasets, such as 3D FRONT (Fu et al., 2021) or Shapenet (Chang et al., 2015), are already supported, making it easier to clutter synthetic scenes with additional objects.


    Access

    Download


    Export, share and cite



    Title :

    BlenderProc2: A Procedural Pipeline for Photorealistic Rendering



    Published in:

    Publication date :

    2023-02-20



    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    Real-time photorealistic stereoscopic rendering of fire [6490A-71]

    Rose, B. M. / McAllister, D. F. / IS & T--the Society for Imaging Science and Technology et al. | British Library Conference Proceedings | 2007


    Learning Accurate Objectness Instance Segmentation from Photorealistic Rendering for Robotic Manipulation

    Li, Siyi / Zhou, Jiaji / Jia, Zhenzhong et al. | Springer Verlag | 2020



    Non-photorealistic rendering as a feedback strategy in virtual reality for rehabilitation

    Quintana, David / Rodríguez Benítez, Antonio / Sbert, Mateu et al. | BASE | 2024

    Free access