Radar-Guided Polynomial Fitting for Metric Depth Estimation
This work addresses the challenge of accurate metric depth estimation for applications like autonomous driving and robotics, offering a more efficient and effective solution compared to existing methods, though it is incremental in refining transformation techniques.
The paper tackles the problem of converting scaleless depth predictions from monocular depth estimation models into accurate metric depth maps by proposing POLAR, a radar-guided polynomial fitting method that corrects misalignments between local depth regions, achieving state-of-the-art performance with average improvements of 24.9% in MAE and 33.2% in RMSE across three datasets.
We propose POLAR, a novel radar-guided depth estimation method that introduces polynomial fitting to efficiently transform scaleless depth predictions from pretrained monocular depth estimation (MDE) models into metric depth maps. Unlike existing approaches that rely on complex architectures or expensive sensors, our method is grounded in a fundamental insight: although MDE models often infer reasonable local depth structure within each object or local region, they may misalign these regions relative to one another, making a linear scale and shift (affine) transformation insufficient given three or more of these regions. To address this limitation, we use polynomial coefficients predicted from cheap, ubiquitous radar data to adaptively adjust depth predictions non-uniformly across depth ranges. In this way, POLAR generalizes beyond affine transformations and is able to correct such misalignments by introducing inflection points. Importantly, our polynomial fitting framework preserves structural consistency through a novel training objective that enforces local monotonicity via first-derivative regularization. POLAR achieves state-of-the-art performance across three datasets, outperforming existing methods by an average of 24.9% in MAE and 33.2% in RMSE, while also achieving state-of-the-art efficiency in terms of latency and computational cost.