Patric Jensfelt
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Patric Jensfelt has a total of 113 co-authors. Here are the top co-authors by number of shared publications:
- Daniel Duberg
Robotics, Perception and Learning, KTH - Royal Institute of Technology, Stockholm Sweden 11364
8 shared publications - John Folkesson
KTH Royal Institute of Technology
8 shared publications - Qingwen Zhang
KTH Royal Institute of Technology
8 shared publications - Leonard Bruns
Division for Robotics, Perception and Learning (RPL KTH Royal Institute of Technology Stockholm Sweden
6 shared publications - Maciej K. Wozniak
KTH Royal Institute of Technology
6 shared publications - Nils Bore
Robotics, Perception and Learning Lab, Royal Institute of Technology, Stockholm, Sweden
4 shared publications
Latest Publications
- 15 September 2026
- Published by Springer Nature
- p. 151-168
SynFlow: Scaling Up LiDAR Scene Flow Estimation with Synthetic Data
- 15 September 2026
- Published by Springer Nature
- p. 151-168
SynFlow: Scaling Up LiDAR Scene Flow Estimation with Synthetic Data
- 14 September 2026
- Published by Springer Nature
- p. 294-314
HilDA: Hierarchical Distillation with Diffusion for Advancing Self-supervised LiDAR Pre-training
- 14 September 2026
- Published by Springer Nature
- p. 294-314
HilDA: Hierarchical Distillation with Diffusion for Advancing Self-supervised LiDAR Pre-training
- Open Access
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (5) , 6400-6407
PRIX: Learning to Plan From Raw Pixels for End-to-End Autonomous Driving
While end-to-end autonomous driving models show promising results, their practical deployment is often hindered by large model sizes, a reliance on expensive LiDAR sensors and computationally intensive BEV feature representations. This limits their scalability, especially for mass-market vehicles equipped only with cameras. To address these challenges, we propose PRIX (Plan from Raw pIXels). Our novel and efficient end-to-end driving architecture operates using only camera data, without explicit BEV representation and forgoing the need for LiDAR. PRIX leverages a visual feature extractor coupled with a generative planning head to predict safe trajectories from raw pixel inputs directly. A core component of our architecture is the Context-aware Recalibration Transformer (CaRT), a novel module designed to effectively enhance multi-level visual features for more robust planning. PRIX achieves SOTA performance on the NavSim-v2 and nuScenes datasets. On NavSim-v1, it also outperforms the majority of multimodal planners and other camera-only approaches on most metrics. Critically, PRIX is significantly more efficient on NavSim-v1, boasting faster inference speeds and a smaller model size. This combination of performance and efficiency makes it a practical solution for real-world deployment.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (5) , 6400-6407
Open AccessPRIX: Learning to Plan From Raw Pixels for End-to-End Autonomous Driving
While end-to-end autonomous driving models show promising results, their practical deployment is often hindered by large model sizes, a reliance on expensive LiDAR sensors and computationally intensive BEV feature representations. This limits their scalability, especially for mass-market vehicles equipped only with cameras. To address these challenges, we propose PRIX (Plan from Raw pIXels). Our novel and efficient end-to-end driving architecture operates using only camera data, without explicit BEV representation and forgoing the need for LiDAR. PRIX leverages a visual feature extractor coupled with a generative planning head to predict safe trajectories from raw pixel inputs directly. A core component of our architecture is the Context-aware Recalibration Transformer (CaRT), a novel module designed to effectively enhance multi-level visual features for more robust planning. PRIX achieves SOTA performance on the NavSim-v2 and nuScenes datasets. On NavSim-v1, it also outperforms the majority of multimodal planners and other camera-only approaches on most metrics. Critically, PRIX is significantly more efficient on NavSim-v1, boasting faster inference speeds and a smaller model size. This combination of performance and efficiency makes it a practical solution for real-world deployment.
- Open Access
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (3) , 3836-3843
DoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance
Accurate 3D scene flow estimation is critical for autonomous systems to navigate dynamic environments safely, but creating the necessary large-scale, manually annotated datasets remains a significant bottleneck for developing robust perception models. Current self-supervised methods struggle to match the performance of fully supervised approaches, especially in challenging long-range and adverse weather scenarios, while supervised methods are not scalable due to their reliance on expensive human labeling. We introduce DoGFlow, a novel self-supervised framework that recovers full 3D object motions for LiDAR scene flow estimation without requiring any manual ground truth annotations. This paper presents our cross-modal label transfer approach, where DoGFlow computes motion labels directly from 4D radar Doppler measurements and transfers them to the LiDAR domain using dynamic-aware association and ambiguity-resolved propagation. On the challenging MAN TruckScenes dataset, DoGFlow substantially outperforms existing self-supervised methods and improves label efficiency by enabling LiDAR backbones to achieve over 90% of fully supervised performance with only 10% of the ground truth data.
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (3) , 3836-3843
Open AccessDoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance
Accurate 3D scene flow estimation is critical for autonomous systems to navigate dynamic environments safely, but creating the necessary large-scale, manually annotated datasets remains a significant bottleneck for developing robust perception models. Current self-supervised methods struggle to match the performance of fully supervised approaches, especially in challenging long-range and adverse weather scenarios, while supervised methods are not scalable due to their reliance on expensive human labeling. We introduce DoGFlow, a novel self-supervised framework that recovers full 3D object motions for LiDAR scene flow estimation without requiring any manual ground truth annotations. This paper presents our cross-modal label transfer approach, where DoGFlow computes motion labels directly from 4D radar Doppler measurements and transfers them to the LiDAR domain using dynamic-aware association and ambiguity-resolved propagation. On the challenging MAN TruckScenes dataset, DoGFlow substantially outperforms existing self-supervised methods and improves label efficiency by enabling LiDAR backbones to achieve over 90% of fully supervised performance with only 10% of the ground truth data.
- Open Access
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (3) , 2530-2537
ProbPer-LiLo: Probabilistic Persistency Modeling for Life-Long Mapping
3D mapping is vital for a broad range of applications that rely on a consistent and accurate representation of the environment. Change is an ever-persistent force in our world and with the evolution of a scene its 3D map becomes outdated. Thus, a mapping framework that can adapt and refine the 3D maps with the changes in the scene is necessary. In this letter, we propose a lifelong mapping framework where map maintenance is based on two objectives including the preservation of static structures and the refinement of the 3D map. To preserve only the static structures, we classify the object’s state and remove the dynamic objects and the quasi-static objects, i.e., objects which temporarily appear static. For classifying the state of objects, we propose a discrete probabilistic solution utilizing a factor graph. Using this classification, we generate static maps from multiple sessions which are used for map refinement. The refinement is based on change detection and map update, leveraging semantic and geometric information. For the evaluation, we collect a multi-campus lifelong dataset as an extension of the MCD datasets from KTH and NTU campuses. The proposed approach is capable of accurately detecting quasi-static objects even in highly dynamic environments. Our system demonstrates state of the art performance in large scale environments. Furthermore, our approach can handle both SLAM-generated and survey-grade maps.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (3) , 2530-2537
Open AccessProbPer-LiLo: Probabilistic Persistency Modeling for Life-Long Mapping
3D mapping is vital for a broad range of applications that rely on a consistent and accurate representation of the environment. Change is an ever-persistent force in our world and with the evolution of a scene its 3D map becomes outdated. Thus, a mapping framework that can adapt and refine the 3D maps with the changes in the scene is necessary. In this letter, we propose a lifelong mapping framework where map maintenance is based on two objectives including the preservation of static structures and the refinement of the 3D map. To preserve only the static structures, we classify the object’s state and remove the dynamic objects and the quasi-static objects, i.e., objects which temporarily appear static. For classifying the state of objects, we propose a discrete probabilistic solution utilizing a factor graph. Using this classification, we generate static maps from multiple sessions which are used for map refinement. The refinement is based on change detection and map update, leveraging semantic and geometric information. For the evaluation, we collect a multi-campus lifelong dataset as an extension of the MCD datasets from KTH and NTU campuses. The proposed approach is capable of accurately detecting quasi-static objects even in highly dynamic environments. Our system demonstrates state of the art performance in large scale environments. Furthermore, our approach can handle both SLAM-generated and survey-grade maps.
- Open Access
Visible Structure Retrieval for Lightweight Image-Based Relocalisation
Accurate camera pose estimation from an image observation in a previously mapped environment is commonly done through structure-based methods: by finding correspondences between 2D keypoints on the image and 3D structure points in the map. In order to make this correspondence search tractable in large scenes, existing pipelines either rely on search heuristics, or perform image retrieval to reduce the search space by comparing the current image to a database of past observations. However, these approaches result in elaborate pipelines or storage requirements that grow with the number of past observations. In this work, we propose a new paradigm for making structure-based relocalisation tractable. Instead of relying on image retrieval or search heuristics, we learn a direct mapping from image observations to the visible scene structure in a compact neural network. Given a query image, a forward pass through our novel visible structure retrieval network allows obtaining the subset of 3D structure points in the map that the image views, thus reducing the search space of 2D-3D correspondences. We show that our proposed method enables performing localisation with an accuracy comparable to the state of the art, while requiring lower computational and storage footprint.
Open AccessVisible Structure Retrieval for Lightweight Image-Based Relocalisation
Accurate camera pose estimation from an image observation in a previously mapped environment is commonly done through structure-based methods: by finding correspondences between 2D keypoints on the image and 3D structure points in the map. In order to make this correspondence search tractable in large scenes, existing pipelines either rely on search heuristics, or perform image retrieval to reduce the search space by comparing the current image to a database of past observations. However, these approaches result in elaborate pipelines or storage requirements that grow with the number of past observations. In this work, we propose a new paradigm for making structure-based relocalisation tractable. Instead of relying on image retrieval or search heuristics, we learn a direct mapping from image observations to the visible scene structure in a compact neural network. Given a query image, a forward pass through our novel visible structure retrieval network allows obtaining the subset of 3D structure points in the map that the image views, thus reducing the search space of 2D-3D correspondences. We show that our proposed method enables performing localisation with an accuracy comparable to the state of the art, while requiring lower computational and storage footprint.
- Open Access
BlendCLIP: Bridging Synthetic and Real Domains for Zero-Shot 3D Object Classification with Multimodal Pretraining
Zero-shot 3D object classification is crucial for real-world applications like autonomous driving, however it is often hindered by a significant domain gap between the synthetic data used for training and the sparse, noisy LiDAR scans encountered in the real-world. Current methods trained solely on synthetic data fail to generalize to outdoor scenes, while those trained only on real data lack the semantic diversity to recognize rare or unseen objects. We introduce BlendCLIP, a multimodal pretraining framework that bridges this synthetic-to-real gap by strategically combining the strengths of both domains. We first propose a pipeline to generate a large-scale dataset of object-level triplets -- consisting of a point cloud, image, and text description -- mined directly from real-world driving data and human annotated 3D boxes. Our core contribution is a curriculum-based data mixing strategy that first grounds the model in the semantically rich synthetic CAD data before progressively adapting it to the specific characteristics of real-world scans. Our experiments show that our approach is highly label-efficient: introducing as few as 1.5\% real-world samples per batch into training boosts zero-shot accuracy on the nuScenes benchmark by 27\%. Consequently, our final model achieves state-of-the-art performance on challenging outdoor datasets like nuScenes and TruckScenes, improving over the best prior method by 19.3\% on nuScenes, while maintaining strong generalization on diverse synthetic benchmarks. Our findings demonstrate that effective domain adaptation, not full-scale real-world annotation, is the key to unlocking robust open-vocabulary 3D perception. Our code and dataset will be released upon acceptance on https://github.com/kesu1/BlendCLIP.
Open AccessBlendCLIP: Bridging Synthetic and Real Domains for Zero-Shot 3D Object Classification with Multimodal Pretraining
Zero-shot 3D object classification is crucial for real-world applications like autonomous driving, however it is often hindered by a significant domain gap between the synthetic data used for training and the sparse, noisy LiDAR scans encountered in the real-world. Current methods trained solely on synthetic data fail to generalize to outdoor scenes, while those trained only on real data lack the semantic diversity to recognize rare or unseen objects. We introduce BlendCLIP, a multimodal pretraining framework that bridges this synthetic-to-real gap by strategically combining the strengths of both domains. We first propose a pipeline to generate a large-scale dataset of object-level triplets -- consisting of a point cloud, image, and text description -- mined directly from real-world driving data and human annotated 3D boxes. Our core contribution is a curriculum-based data mixing strategy that first grounds the model in the semantically rich synthetic CAD data before progressively adapting it to the specific characteristics of real-world scans. Our experiments show that our approach is highly label-efficient: introducing as few as 1.5\% real-world samples per batch into training boosts zero-shot accuracy on the nuScenes benchmark by 27\%. Consequently, our final model achieves state-of-the-art performance on challenging outdoor datasets like nuScenes and TruckScenes, improving over the best prior method by 19.3\% on nuScenes, while maintaining strong generalization on diverse synthetic benchmarks. Our findings demonstrate that effective domain adaptation, not full-scale real-world annotation, is the key to unlocking robust open-vocabulary 3D perception. Our code and dataset will be released upon acceptance on https://github.com/kesu1/BlendCLIP.
- Open Access
- 8 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 41 (15523098) , 5896-5911
HiMo: High-Speed Objects Motion Compensation in Point Clouds
LiDAR point cloud is essential for autonomous vehicles, but motion distortions from dynamic objects degrade the data quality. While previous work has considered distortions caused by ego motion, distortions caused by other moving objects remain largely overlooked, leading to errors in object shape and position. This distortion is particularly pronounced in high-speed environments such as highways and in multi-LiDAR configurations, a common setup for heavy vehicles. To address this challenge, we introduce HiMo, a pipeline that repurposes scene flow estimation for non-ego motion compensation, correcting the representation of dynamic objects in point clouds. During the development of HiMo, we observed that existing self-supervised scene flow estimators often produce degenerate or inconsistent estimates under high-speed distortion. We further propose SeFlow++, a real-time scene flow estimator that achieves state-of-the-art performance on both scene flow and motion compensation. Since well-established motion distortion metrics are absent in the literature, we introduce two evaluation metrics: compensation accuracy at a point level and shape similarity of objects. We validate HiMo through extensive experiments on Argoverse 2, ZOD and a newly collected real-world dataset featuring highway driving and multi-LiDAR-equipped heavy vehicles. Our findings show that HiMo improves the geometric consistency and visual fidelity of dynamic objects in LiDAR point clouds, benefiting downstream tasks such as semantic segmentation and 3D detection. See https://kin-zhang.github.io/HiMo for more details.
- 8 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 41 (15523098) , 5896-5911
Open AccessHiMo: High-Speed Objects Motion Compensation in Point Clouds
LiDAR point cloud is essential for autonomous vehicles, but motion distortions from dynamic objects degrade the data quality. While previous work has considered distortions caused by ego motion, distortions caused by other moving objects remain largely overlooked, leading to errors in object shape and position. This distortion is particularly pronounced in high-speed environments such as highways and in multi-LiDAR configurations, a common setup for heavy vehicles. To address this challenge, we introduce HiMo, a pipeline that repurposes scene flow estimation for non-ego motion compensation, correcting the representation of dynamic objects in point clouds. During the development of HiMo, we observed that existing self-supervised scene flow estimators often produce degenerate or inconsistent estimates under high-speed distortion. We further propose SeFlow++, a real-time scene flow estimator that achieves state-of-the-art performance on both scene flow and motion compensation. Since well-established motion distortion metrics are absent in the literature, we introduce two evaluation metrics: compensation accuracy at a point level and shape similarity of objects. We validate HiMo through extensive experiments on Argoverse 2, ZOD and a newly collected real-world dataset featuring highway driving and multi-LiDAR-equipped heavy vehicles. Our findings show that HiMo improves the geometric consistency and visual fidelity of dynamic objects in LiDAR point clouds, benefiting downstream tasks such as semantic segmentation and 3D detection. See https://kin-zhang.github.io/HiMo for more details.
- 16 June 2025
- Published by Springer Nature
- p. 180-195
Quantifying Epistemic Uncertainty in Absolute Pose Regression
- 16 June 2025
- Published by Springer Nature
- p. 180-195
Quantifying Epistemic Uncertainty in Absolute Pose Regression
- Open Access
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Intelligent Transportation Systems
- Vol. 6, 1-10
Transitional Grid Maps: Joint Modeling of Static and Dynamic Occupancy
Autonomous agents rely on sensor data to construct representations of their environments, essential for predicting future events and planning their actions. However, sensor measurements suffer from limited range, occlusions, and sensor noise. These challenges become more evident in highly dynamic environments.This work proposes a probabilistic framework to jointly infer which parts of an environment are statically and which parts are dynamically occupied.We formulate the problem as a Bayesian network and introduce minimal assumptions that significantly reduce the complexity of the problem. Based on those, we derive Transitional Grid Maps (TGMs), an efficient analytical solution.Using real data, we demonstrate how this approach produces better maps than the state-of-the-art by keeping track of both static and dynamic elements and, as a side effect, can help improve existing SLAM algorithms.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Intelligent Transportation Systems
- Vol. 6, 1-10
Open AccessTransitional Grid Maps: Joint Modeling of Static and Dynamic Occupancy
Autonomous agents rely on sensor data to construct representations of their environments, essential for predicting future events and planning their actions. However, sensor measurements suffer from limited range, occlusions, and sensor noise. These challenges become more evident in highly dynamic environments.This work proposes a probabilistic framework to jointly infer which parts of an environment are statically and which parts are dynamically occupied.We formulate the problem as a Bayesian network and introduce minimal assumptions that significantly reduce the complexity of the problem. Based on those, we derive Transitional Grid Maps (TGMs), an efficient analytical solution.Using real data, we demonstrate how this approach produces better maps than the state-of-the-art by keeping track of both static and dynamic elements and, as a side effect, can help improve existing SLAM algorithms.