Estimating Small Differences in Car-Pose from Orbits
This work addresses the problem of fine-grained pose estimation for objects like cars, which is incremental as it builds on existing pose estimation methods by focusing on relative differences.
The paper tackles the challenge of distinguishing small pose differences and symmetries in objects by proposing a group-theoretic method that estimates pose differences relative to another pose, rather than predicting absolute pose, and demonstrates its effectiveness on cars where subtle pose distinctions are crucial.
Distinction among nearby poses and among symmetries of an object is challenging. In this paper, we propose a unified, group-theoretic approach to tackle both. Different from existing works which directly predict absolute pose, our method measures the pose of an object relative to another pose, i.e., the pose difference. The proposed method generates the complete orbit of an object from a single view of the object with respect to the subgroup of SO(3) of rotations around the z-axis, and compares the orbit of the object with another orbit using a novel orbit metric to estimate the pose difference. The generated orbit in the latent space records all the differences in pose in the original observational space, and as a result, the method is capable of finding subtle differences in pose. We demonstrate the effectiveness of the proposed method on cars, where identifying the subtle pose differences is vital.