RAD: Retrieval-Augmented Monocular Metric Depth Estimation for Underrepresented Classes

Monocular Metric Depth Estimation (MMDE) is essential for physically intelligent systems, yet accurate depth estimation for underrepresented classes in complex scenes remains a persistent challenge. To address this, we propose RAD, a retrieval-augmented framework that approximates the benefits of multi-view stereo by utilizing retrieved neighbors as structural geometric proxies. Our method first employs an uncertainty-aware retrieval mechanism to identify low-confidence regions in the input and retrieve RGB-D context samples containing semantically similar content. We then process both the input and retrieved context via a dual-stream network and fuse them using a matched cross-attention module, which transfers geometric information only at reliable point correspondences. Evaluations on NYU Depth v2, KITTI, and Cityscapes demonstrate that RAD significantly outperforms state-of-the-art baselines on underrepresented classes, reducing relative absolute error by 29.2% on NYU Depth v2, 13.3% on KITTI, and 7.2% on Cityscapes, while maintaining competitive performance on standard in-domain benchmarks.

Polarimetric Imaging for Perception

Autonomous driving and advanced driver-assistance systems rely on a set of sensors and algorithms to perform the appropriate actions and provide alerts as a function of the driving scene. Typically, the sensors include color cameras, radar, lidar and ultrasonic sensors. Strikingly however, although light polarization is a fundamental property of light, it is seldom harnessed for perception tasks. In this work we analyze the potential for improvement in perception tasks when using an RGB-polarimetric camera, as compared to an RGB camera. We examine monocular depth estimation and free space detection during the middle of the day, when polarization is independent of subject heading, and show that a quantifiable improvement can be achieved for both of them using state-of-the-art deep neural networks, with a minimum of architectural changes. We also present a new dataset composed of RGB-polarimetric images, lidar scans, GNSS / IMU readings and free space segmentations that further supports developing perception algorithms that take advantage of light polarization.

Local Variation as a Statistical Hypothesis Test

The goal of image oversegmentation is to divide an image into several pieces, each of which should ideally be part of an object. One of the simplest and yet most effective oversegmentation algorithms is known as local variation (LV) (Felzenszwalb and Huttenlocher 2004). In this work, we study this algorithm and show that algorithms similar to LV can be devised by applying different statistical models and decisions, thus providing further theoretical justification and a well-founded explanation for the unexpected high performance of the LV approach. Some of these algorithms are based on statistics of natural images and on a hypothesis testing decision; we denote these algorithms probabilistic local variation (pLV). The best pLV algorithm, which relies on censored estimation, presents state-of-the-art results while keeping the same computational complexity of the LV algorithm.