Conditional Variational Autoencoders for Probabilistic Pose Regression
Not on the map. Why?
Abstract (arXiv)
Robots rely on visual relocalization to estimate their pose from camera images when they lose track. One of the challenges in visual relocalization is repetitive structures in the operation environment of the robot. This calls for probabilistic methods that support multiple hypotheses for robot's pose. We propose such a probabilistic method to predict the posterior distribution of camera poses given an observed image. Our proposed training strategy results in a generative model of camera poses given an image, which can be used to draw samples from the pose posterior distribution. Our method is streamlined and well-founded in theory and outperforms existing methods on localization in presence of ambiguities.
Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Builds on 33 of its 48 references that are in the map. First posted on arXiv 7 Oct 2024.
Builds on
33 of 48 references.
References · 33
- 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,712
- 2015Variational Inference with Normalizing Flowscited by 2,163
- 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,195
- 2019An Introduction to Variational Autoencoderscited by 998
- 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 935
- 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 557
- 2016Tutorial on Variational Autoencoderscited by 404
- 2017Toward Multimodal Image-to-Image Translationcited by 295
- 2021Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognitioncited by 236
- 2017Geometric Loss Functions for Camera Pose Regression with Deep Learningcited by 224
- 2019Understanding the Limitations of CNN-based Absolute Camera Pose Regressioncited by 177
- 2020Visual Camera Re-Localization from RGB and RGB-D Images Using DSACcited by 176
21 more
- 2017Geometry-Aware Learning of Maps for Camera Localizationcited by 161
- 2021Probabilistic Modeling for Human Mesh Recoverycited by 152
- 2017Learning Less is More - 6D Camera Localization via 3D Surface Regressioncited by 146
- 2016Image-based localization using LSTMs for structured feature correlationcited by 136
- 2015Modelling Uncertainty in Deep Learning for Camera Relocalizationcited by 133
- 2021LENS: Localization enhanced by NeRF synthesiscited by 117
- 2021Learning Multi-Scene Absolute Pose Regression with Transformerscited by 108
- 2022DiffPose: Toward More Reliable 3D Pose Estimationcited by 100
- 2019AtLoc: Attention Guided Camera Localizationcited by 92
- 2016An Uncertain Future: Forecasting from Static Images using Variational Autoencoderscited by 86
- 2022DFNet: Enhance Absolute Pose Regression with Direct Feature Matchingcited by 82
- 2017Image-based Localization using Hourglass Networkscited by 63
- 2021Direct-PoseNet: Absolute Pose Regression with Photometric Consistencycited by 61
- 2023GenPose: Generative Category-level Object Pose Estimation via Diffusion Modelscited by 48
- 2020Deep Bingham Networks: Dealing with Uncertainty and Ambiguity in Pose Estimationcited by 43
- 2021CoordiNet: uncertainty-aware pose regressor for reliable vehicle localizationcited by 34
- 2021Reassessing the Limitations of CNN Methods for Camera Pose Regressioncited by 19
- 2022Leveraging Equivariant Features for Absolute Pose Regressioncited by 17
- 2021A Multi-Hypothesis Approach to Pose Ambiguity in Object-Based SLAMcited by 12
- 2023A Probabilistic Framework for Visual Localization in Ambiguous Scenescited by 12
- 2023CaLDiff: Camera Localization in NeRF via Pose Diffusioncited by 2
Built on by
Citing papers per year: 2025: 2.
2 papers.