BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes)
<p>Robots require high-quality <em>maps</em>—internal representations of their operating workspace—to localise, path plan, and perceive their environment. Until recently, these maps were restricted to sparse, 2D representations due to computational, memory, and sensor limitations....
| 1. autor: | |
|---|---|
| Kolejni autorzy: | |
| Format: | Praca dyplomowa |
| Język: | English |
| Wydane: |
2017
|
| Hasła przedmiotowe: |
| _version_ | 1826317371519795200 |
|---|---|
| author | Tanner, M |
| author2 | Newman, P |
| author_facet | Newman, P Tanner, M |
| author_sort | Tanner, M |
| collection | OXFORD |
| description | <p>Robots require high-quality <em>maps</em>—internal representations of their operating workspace—to localise, path plan, and perceive their environment. Until recently, these maps were restricted to sparse, 2D representations due to computational, memory, and sensor limitations. With the widespread adoption of high-quality sensors and graphics processors for parallel processing, these restrictions no longer apply: dense 3D maps are feasible to compute in real time (i.e., at the input sensor’s frame rate).</p> <p>This thesis presents the theory and system to create large-scale dense 3D maps (i.e., reconstruct continuous surface models) using only sensors found on modern autonomous automobiles: 2D laser, 3D laser, and cameras. In contrast to active RGB-D cameras, passive cameras produce noisy surface observations and must be regularised in both 2D and 3D to create accurate reconstructions. Unfortunately, straight-forward application of 3D regularisation causes undesired surface interpolation and extrapolation in regions unexplored by the robot. We propose a method to overcome this challenge by informing the regulariser of the specific subsets of 3D surfaces upon which to operate. When combined with a compressed voxel grid data structure, we demonstrate our system fusing data from both laser and camera sensors to reconstruct 7.3 km of urban environments. We evaluate the quantitative performance of our proposed method through the use of synthetic and real-world datasets—including datasets from Stanford's Burghers of Calais, University of Oxford's RobotCar, University of Oxford’s Dense Reconstruction, and Karlsruhe Institute of Technology’s KITTI—compared to ground-truth laser data. With only stereo camera inputs, our regulariser reduces the 3D reconstruction metric error between 27% to 36% with a final median accuracy ranging between 4 cm to 8 cm.</p> <p>Furthermore, by augmenting our system with object detection, we remove ephemeral objects (e.g., automobiles, bicycles, and pedestrians) from the input sensor data and target our regulariser to interpolate the occluded urban surfaces. Augmented with Kernel Conditional Density Estimation, our regulariser creates reconstructions with median errors between 5.64 cm and 9.24 cm.</p> <p>Finally, we present a machine-learning pipeline that learns, in an automatic fashion, to recognise the errors in dense reconstructions. Our system trains on image and laser data from a 3.8 km urban sequence. Using a separate 2.2 km urban sequence, our pipeline consistently identifies error-prone regions in the image-based dense reconstruction.</p> |
| first_indexed | 2024-03-06T19:18:13Z |
| format | Thesis |
| id | oxford-uuid:1928c996-d913-4d7e-8ca5-cf247f90aa0f |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:37:24Z |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oxford-uuid:1928c996-d913-4d7e-8ca5-cf247f90aa0f2025-02-11T16:11:04ZBOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes)Thesishttp://purl.org/coar/resource_type/c_db06uuid:1928c996-d913-4d7e-8ca5-cf247f90aa0foptimisationautonomous vehiclesroboticsmachine learningartificial intelligencemappingEnglishORA Deposit2017Tanner, MNewman, P<p>Robots require high-quality <em>maps</em>—internal representations of their operating workspace—to localise, path plan, and perceive their environment. Until recently, these maps were restricted to sparse, 2D representations due to computational, memory, and sensor limitations. With the widespread adoption of high-quality sensors and graphics processors for parallel processing, these restrictions no longer apply: dense 3D maps are feasible to compute in real time (i.e., at the input sensor’s frame rate).</p> <p>This thesis presents the theory and system to create large-scale dense 3D maps (i.e., reconstruct continuous surface models) using only sensors found on modern autonomous automobiles: 2D laser, 3D laser, and cameras. In contrast to active RGB-D cameras, passive cameras produce noisy surface observations and must be regularised in both 2D and 3D to create accurate reconstructions. Unfortunately, straight-forward application of 3D regularisation causes undesired surface interpolation and extrapolation in regions unexplored by the robot. We propose a method to overcome this challenge by informing the regulariser of the specific subsets of 3D surfaces upon which to operate. When combined with a compressed voxel grid data structure, we demonstrate our system fusing data from both laser and camera sensors to reconstruct 7.3 km of urban environments. We evaluate the quantitative performance of our proposed method through the use of synthetic and real-world datasets—including datasets from Stanford's Burghers of Calais, University of Oxford's RobotCar, University of Oxford’s Dense Reconstruction, and Karlsruhe Institute of Technology’s KITTI—compared to ground-truth laser data. With only stereo camera inputs, our regulariser reduces the 3D reconstruction metric error between 27% to 36% with a final median accuracy ranging between 4 cm to 8 cm.</p> <p>Furthermore, by augmenting our system with object detection, we remove ephemeral objects (e.g., automobiles, bicycles, and pedestrians) from the input sensor data and target our regulariser to interpolate the occluded urban surfaces. Augmented with Kernel Conditional Density Estimation, our regulariser creates reconstructions with median errors between 5.64 cm and 9.24 cm.</p> <p>Finally, we present a machine-learning pipeline that learns, in an automatic fashion, to recognise the errors in dense reconstructions. Our system trains on image and laser data from a 3.8 km urban sequence. Using a separate 2.2 km urban sequence, our pipeline consistently identifies error-prone regions in the image-based dense reconstruction.</p> |
| spellingShingle | optimisation autonomous vehicles robotics machine learning artificial intelligence mapping Tanner, M BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title | BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title_full | BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title_fullStr | BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title_full_unstemmed | BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title_short | BOR2G: Building Optimal Regularised Reconstructions with GPUs (in cubes) |
| title_sort | bor2g building optimal regularised reconstructions with gpus in cubes |
| topic | optimisation autonomous vehicles robotics machine learning artificial intelligence mapping |
| work_keys_str_mv | AT tannerm bor2gbuildingoptimalregularisedreconstructionswithgpusincubes |