Deep Exemplar 2D-3D Detection by Adapting from Real to Rendered Views
It addresses object detection challenges in computer vision by enabling better alignment between real and synthetic data, though it is incremental as it builds on existing deep learning pipelines.
The paper tackles 2D-3D exemplar detection by adapting features from real images to CAD rendered views using a CNN, achieving improved accuracy and speed on tasks like instance detection on the IKEA dataset and outperforming prior work for chair detection on Pascal VOC.
This paper presents an end-to-end convolutional neural network (CNN) for 2D-3D exemplar detection. We demonstrate that the ability to adapt the features of natural images to better align with those of CAD rendered views is critical to the success of our technique. We show that the adaptation can be learned by compositing rendered views of textured object models on natural images. Our approach can be naturally incorporated into a CNN detection pipeline and extends the accuracy and speed benefits from recent advances in deep learning to 2D-3D exemplar detection. We applied our method to two tasks: instance detection, where we evaluated on the IKEA dataset, and object category detection, where we out-perform Aubry et al. for "chair" detection on a subset of the Pascal VOC dataset.