Depth map estimation is crucial for a wide range of applications. Unfortunately, it often presents missing or unreliable data. The objective of depth completion is to fill in the “holes” in a depth map by propagating the depth information using guidance from other sources of information, such as color. Nowadays, classical image processing methods have been outperformed by deep learning techniques. Nevertheless, these approaches require a significantly large number of images and enormous computing power for training. This fact limits their usability and makes them not the best solution in some resource-constrained environments. Therefore, this paper investigates three simple hybrid models for depth completion. We explore a hybrid pipeline that combines a very efficient and powerful interpolator (infinity Laplacian or AMLE) and a series of convolutional stages. The contributions of this article are (i) the use a Texture+Structuredecomposition as a pre-filter stage; (ii) an objective evaluation with three different approaches using KITTI and NYU_V2 data sets; (iii) the use of an anisotropic metric as a mechanism to improve interpolation; and iv) the inclusion of an ablation test. The main conclusions of this work are that using an anisotropic metric improves model performance, and the ablation test demonstrates that the model’s final stage is a critical component in the pipeline; its suppression leads to an approximate 4% increase in $MSE$. We also show that our model outperforms state-of-the-art alternatives with similar levels of complexity.

Scene depth information is a key component of any robotic mobile application. Range sensors, such as LiDAR, sonar, or radar, capture depth data of a scene. However, the data captured by these sensors frequently presents missing regions or information with a low confidence level. These missing regions in the depth data could be large areas without information, making it difficult to make decisions, for instance, for an autonomous vehicle. Recovering depth data has become a primary activity for computer vision applications. This work proposes and evaluates an interpolation model to infer dense depth maps from a Lab color space reference picture and an incomplete-depth image embedded in a completion pipeline. The complete proposal pipeline comprises convolutional layers and a convex combination of the infinity Laplacian and non-local means model. The proposed model infers dense depth maps by considering depth data and utilizing clues from a color picture of the scene, along with a metric for computing differences between two pixels. The work contributes (i) the convex combination of the two models to interpolate the data, and (ii) the proposal of a class of function suitable for balancing between different models. The obtained results show that the model outperforms similar models in the KITTI dataset and outperforms our previous implementation in the NYU_v2 dataset, dropping the MSE by 34.86%, 3.35%, and 34.42% for 4×, 8×, 16× upsampling tasks, respectively.

In this paper, we present a comprehensive approach for depth map completion using a variable pipeline model. The pipeline is designed to be flexible, and its structure is dynamically determined during the training phase by minimizing a fitness function. To achieve this objective, we used the Particle Swarm Optimization (PSO) technique, which efficiently ex-plores the solution space to optimize key parameters. The pipeline itself consists of three main stages: a convolutional stage (Lab Convolution), an Interpolation Model (IM), and a post-Convolutional Stage (SC2). During training, our model automatically adjusts several aspects, including the selection of filter parameters, interpolator settings, and the number of filters used at each stage. Moreover, the pipeline is capable of interchanging the execution order between the stages, either processing in the sequence of Lab Convolution-IM-SC2 or SC2-IM-Lab Convolution. This flexible structure allows for the discovery of better configurations that significantly im-prove the performance of depth map completion tasks. Our approach demonstrates superior results by adapting to the specific characteristics of the input data.