SS-Pose: Self-Supervised 6-D Object Pose Representation Learning Without Rendering

Object pose estimation has extensive applications in various industrial scenarios. However, the heavy reliance on dense 6-D annotation and textured object models has become a significant obstacle to the widespread industrial application of 6-D object pose estimation methods. In this work, we present <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SS-Pose</i>, a self-supervised learning framework for estimating 6-D object poses without annotated 6-D data and textured model. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SS-Pose</i> proposes the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">coordinate system datum reinitializer</i> stage to dynamically establish a sequence-level pose representation datum, and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">temporal–spatial constraint resolver</i> module to obtain the self-supervised learning target through interframe constraints. We introduce a one-shot <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cross-coordinate transformation</i> that establishes the relationship between the 6-D representation and the object poses, which can be further utilized in real-world tasks. We evaluated the proposed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SS-Pose</i> on the challenging YCB-Video dataset and texture-less T-LESS dataset. Our approach achieves competitive performance with significantly lower data dependency, making it suitable for visual perception in industrial applications.