ConvPoseCNN2: Prediction and Refinement of Dense 6D Object Poses

TL;DR

ConvPoseCNN2 is a fully convolutional network that densely predicts 6D object poses, offering improved spatial resolution, efficiency, and refinement capabilities, achieving state-of-the-art accuracy on challenging datasets.

Contribution

It extends PoseCNN with a dense, fully convolutional architecture and introduces an iterative refinement module for enhanced pose prediction accuracy.

Findings

Achieves comparable accuracy to PoseCNN on YCB-Video dataset

Reduces model parameters and inference time

Improves pose predictions through iterative refinement

Abstract

Object pose estimation is a key perceptual capability in robotics. We propose a fully-convolutional extension of the PoseCNN method, which densely predicts object translations and orientations. This has several advantages such as improving the spatial resolution of the orientation predictions -- useful in highly-cluttered arrangements, significant reduction in parameters by avoiding full connectivity, and fast inference. We propose and discuss several aggregation methods for dense orientation predictions that can be applied as a post-processing step, such as averaging and clustering techniques. We demonstrate that our method achieves the same accuracy as PoseCNN on the challenging YCB-Video dataset and provide a detailed ablation study of several variants of our method. Finally, we demonstrate that the model can be further improved by inserting an iterative refinement module into the middle of the network, which enforces consistency of the prediction.