The problem of 3D layout recovery in indoor scenes has been a core research topic for over a decade. However, there are still several major challenges that remain unsolved. Among the most relevant ones, a major part of the state-of-the-art methods make implicit or explicit assumptions on the scenes -e.g. box-shaped or Manhattan layouts. Also, current methods are computationally expensive and not suitable for real-time applications like robot navigation and AR/VR. In this work we present CFL (Corners for Layout), the first end-to-end model that predicts layout corners for 3D layout recovery on 360° images. Our experimental results show that we outperform the state of the art, making less assumptions on the scene than other works, and with lower cost. We also show that our model generalizes better to camera position variations than conventional approaches by using EquiConvs, a convolution applied directly on the spherical projection and hence invariant to the equirectangular distortions.

A novel method to detect stairs with a RGB-D camera is proposed.We get the fully measured model with pose to validate and help in navigation.The system is designed to be wearable and aimed to assist the visually impaired.Visual odometry is computed to enhance the navigation system in video sequences.On-line stair measurements are used to correct the drift of the visual odometry. Stairs are one of the most common structures present in human-made scenarios, but also one of the most dangerous for those with vision problems. In this work we propose a complete method to detect, locate and parametrise stairs with a wearable RGB-D camera. Our algorithm uses the depth data to determine if the horizontal planes in the scene are valid steps of a staircase judging their dimensions and relative positions. As a result we obtain a scaled model of the staircase with the spatial location and orientation with respect to the subject. The visual odometry is also estimated to continuously recover the current position and orientation of the user while moving. This enhances the system giving the ability to come back to previously detected features and providing location awareness of the user during the climb. Simultaneously, the detection of the staircase during the traversal is used to correct the drift of the visual odometry. A comparison of results of the stair detection with other state-of-the-art algorithms was performed using public dataset. Additional experiments have also been carried out, recording our own natural scenes with a chest-mounted RGB-D camera in indoor scenarios. The algorithm is robust enough to work in real-time and even under partial occlusions of the stair.

In this paper, we propose a novel procedure for 3D layout recovery of indoor scenes from single 360 degrees panoramic images. With such images, all scene is seen at once, allowing to recover closed geometries. Our method combines strategically the accuracy provided by geometric reasoning (lines and vanishing points) with the higher level of data abstraction and pattern recognition achieved by deep learning techniques (edge and normal maps). Thus, we extract structural corners from which we generate layout hypotheses of the room assuming Manhattan world. The best layout model is selected, achieving good performance on both simple rooms (box-type) and complex shaped rooms (with more than four walls). Experiments of the proposed approach are conducted within two public datasets, SUN360 and Stanford (2D-3D-S) demonstrating the advantages of estimating layouts by combining geometry and deep learning and the effectiveness of our proposal with respect to the state of the art.

We present a leaderless distributed controller for driving a multirobot team in a planar workspace toward an affine formation, i.e., an affine transformation of a nominal configuration. Our central idea is to organize the team in interlaced modules. Specifically, we define a module as a group of four robots, and we consider two given modules interlaced if they have three robots in common. For each module we define a cost, based on least-squares affine alignment between the positions of the four robots in the current and nominal configurations. Our strategy for formation achievement is to make the robots descend along the gradient of the sum of module costs. Based on this strategy, we propose a distributed control law considering the single-integrator dynamic model. Our main contribution is that the proposed modular approach allows design and reconfiguration to be done locally, i.e., involving only the robots that belong to the modules being designed or reconfigured. We present a formal stability study and an implementation algorithm. To motivate the practical interest of the proposed approach, we illustrate its usage in a multitarget enclosing and tracking scenario. The approach is experimentally validated using simulations and tests with physical unicycle robots.