Search Results for author: Patricio A. Vela

Found 24 papers, 13 papers with code

Planning with Sequence Models through Iterative Energy Minimization

no code implementations28 Mar 2023 Hongyi Chen, Yilun Du, Yiye Chen, Joshua Tenenbaum, Patricio A. Vela

In this paper, we suggest an approach towards integrating planning with sequence models based on the idea of iterative energy minimization, and illustrate how such a procedure leads to improved RL performance across different tasks.

Language Modelling Reinforcement Learning (RL)

Parallel Inversion of Neural Radiance Fields for Robust Pose Estimation

1 code implementation18 Oct 2022 Yunzhi Lin, Thomas Müller, Jonathan Tremblay, Bowen Wen, Stephen Tyree, Alex Evans, Patricio A. Vela, Stan Birchfield

We present a parallelized optimization method based on fast Neural Radiance Fields (NeRF) for estimating 6-DoF pose of a camera with respect to an object or scene.

Pose Estimation

Geometry of Radial Basis Neural Networks for Safety Biased Approximation of Unsafe Regions

no code implementations11 Oct 2022 Ahmad Abuaish, Mohit Srinivasan, Patricio A. Vela

This manuscript describes the specific geometry of the neural network used for zeroing barrier function synthesis, and shows how the network provides the necessary representation for splitting the state space into safe and unsafe regions.

Keypoint-Based Category-Level Object Pose Tracking from an RGB Sequence with Uncertainty Estimation

1 code implementation23 May 2022 Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield

We propose a single-stage, category-level 6-DoF pose estimation algorithm that simultaneously detects and tracks instances of objects within a known category.

Pose Estimation Pose Tracking

Single-Stage Keypoint-Based Category-Level Object Pose Estimation from an RGB Image

1 code implementation13 Sep 2021 Yunzhi Lin, Jonathan Tremblay, Stephen Tyree, Patricio A. Vela, Stan Birchfield

Prior work on 6-DoF object pose estimation has largely focused on instance-level processing, in which a textured CAD model is available for each object being detected.

Object object-detection +2

GKNet: grasp keypoint network for grasp candidates detection

no code implementations16 Jun 2021 Ruinian Xu, Fu-Jen Chu, Patricio A. Vela

Decreasing the detection difficulty by grouping keypoints into pairs boosts performance.

Keypoint Detection

Good Graph to Optimize: Cost-Effective, Budget-Aware Bundle Adjustment in Visual SLAM

2 code implementations23 Aug 2020 Yipu Zhao, Justin S. Smith, Patricio A. Vela

The cost-efficiency of visual(-inertial) SLAM (VSLAM) is a critical characteristic of resource-limited applications.

Good Feature Matching: Towards Accurate, Robust VO/VSLAM with Low Latency

2 code implementations3 Jan 2020 Yipu Zhao, Patricio A. Vela

Analysis of state-of-the-art VO/VSLAM system exposes a gap in balancing performance (accuracy & robustness) and efficiency (latency).

Using Synthetic Data and Deep Networks to Recognize Primitive Shapes for Object Grasping

1 code implementation12 Sep 2019 Yunzhi Lin, Chao Tang, Fu-Jen Chu, Patricio A. Vela

Each primitive shape is designed with parametrized grasp families, permitting the pipeline to identify multiple grasp candidates per shape primitive region.

Recognizing Object Affordances to Support Scene Reasoning for Manipulation Tasks

1 code implementation12 Sep 2019 Fu-Jen Chu, Ruinian Xu, Chao Tang, Patricio A. Vela

Unfortunately, the top performing affordance recognition methods use object category priors to boost the accuracy of affordance detection and segmentation.

Affordance Detection Affordance Recognition +4

Characterizing SLAM Benchmarks and Methods for the Robust Perception Age

1 code implementation19 May 2019 Wenkai Ye, Yipu Zhao, Patricio A. Vela

The ad-hoc creation of these benchmarks does not necessarily illuminate the particular weak points of a SLAM algorithm when performance is evaluated.

Benchmarking

Low-latency Visual SLAM with Appearance-Enhanced Local Map Building

no code implementations19 May 2019 Yipu Zhao, Wenkai Ye, Patricio A. Vela

This paper describes an enhancement to co-visibility local map building by incorporating a strong appearance prior, which leads to a more compact local map and latency reduction in downstream data association.

Pose Estimation Pose Tracking

Good Feature Selection for Least Squares Pose Optimization in VO/VSLAM

no code implementations19 May 2019 Yipu Zhao, Patricio A. Vela

This paper aims to select features that contribute most to the pose estimation in VO/VSLAM.

Robotics

Good Line Cutting: towards Accurate Pose Tracking of Line-assisted VO/VSLAM

no code implementations ECCV 2018 Yipu Zhao, Patricio A. Vela

This paper tackles a problem in line-assisted VO/VSLAM: accurately solving the least squares pose optimization with unreliable 3D line input.

Pose Estimation Pose Tracking

Deep Grasp: Detection and Localization of Grasps with Deep Neural Networks

4 code implementations1 Feb 2018 Fu-Jen Chu, Ruinian Xu, Patricio A. Vela

By defining the learning problem to be classification with null hypothesis competition instead of regression, the deep neural network with RGB-D image input predicts multiple grasp candidates for a single object or multiple objects, in a single shot.

Robotics

Learning Binary Features Online from Motion Dynamics for Incremental Loop-Closure Detection and Place Recognition

no code implementations15 Jan 2016 Guangcong Zhang, Mason J. Lilly, Patricio A. Vela

A codeword for bag-of-words models is generated by packaging the learned descriptor and mask, with a masked Hamming distance defined to measure the distance between two codewords.

Loop Closure Detection

Reduced-Set Kernel Principal Components Analysis for Improving the Training and Execution Speed of Kernel Machines

no code implementations26 Jul 2015 Hassan A. Kingravi, Patricio A. Vela, Alexandar Gray

This paper presents a practical, and theoretically well-founded, approach to improve the speed of kernel manifold learning algorithms relying on spectral decomposition.

Good Features to Track for Visual SLAM

no code implementations CVPR 2015 Guangcong Zhang, Patricio A. Vela

Efficient computation strategies for the observability indices are described based on incremental singular value decomposition (SVD) and greedy selection for the temporal and instantaneous observability indices, respectively.

Cannot find the paper you are looking for? You can Submit a new open access paper.