Per-Erik Forssén

Pavlo Melnyk, Cuong Le, Urs Waldmann et al.

Estimating 3D from 2D is one of the central tasks in computer vision. In this work, we consider the monocular setting, i.e. single-view input, for 3D human pose estimation (HPE). Here, the task is to predict a 3D point set of human skeletal joints from a single 2D input image. While by definition this is an ill-posed problem, recent work has presented methods that solve it with up to several-centimetre error. Typically, these methods employ a two-step approach, where the first step is to detect the 2D skeletal joints in the input image, followed by the step of 2D-to-3D lifting. We find that common lifting models fail when encountering a rotated input. We argue that learning a single human pose along with its in-plane rotations is considerably easier and more geometrically grounded than directly learning a point-to-point mapping. Furthermore, our intuition is that endowing the model with the notion of rotation equivariance without explicitly constraining its parameter space should lead to a more straightforward learning process than one with equivariance by design. Utilising the common HPE benchmarks, we confirm that the 2D rotation equivariance per se improves the model performance on human poses akin to rotations in the image plane, and can be efficiently and straightforwardly learned by augmentation, outperforming state-of-the-art equivariant-by-design methods.

Bao-Long Tran, Per-Erik Forssén, Fredrik Viksten

In this paper, we introduce a challenging task: extracting a fundamental matrix from a single motion blurred image. For a camera moving in 3D during exposure, the smear paths in the blurry image contain cues and constraints on this motion. We demonstrate the feasibility of establishing correspondences between two time instances within the camera exposure window, and that these can be used to robustly infer a fundamental matrix, which summarizes the motion of the camera during the exposure time. The inferred fundamental matrix is unique up to a transpose, corresponding to an ambiguity of the direction of time. Due to this per-smear ambiguity, classic methods, such as the 8-point algorithm, are no longer usable. The proposed method modifies the estimation to work on time-direction ambiguous correspondences. To improve the robustness of the fundamental matrix estimation, we also propose to incorporate an uncertainty measurement in smear pattern prediction and use it in the sampling process of the estimator. Experiments on synthetic and real-world motion-blur datasets demonstrate that our approach is able to estimate the fundamental matrix encoding the 3D camera motion, from single frames. Practical applicability is demonstrated on the downstream task of motion segmentation.

Zahra Gharaee, Felix Järemo Lawin, Per-Erik Forssén

This paper describes recent developments in object specific pose and shape prediction from single images. The main contribution is a new approach to camera pose prediction by self-supervised learning of keypoints corresponding to locations on a category specific deformable shape. We designed a network to generate a proxy ground-truth heatmap from a set of keypoints distributed all over the category-specific mean shape, where each is represented by a unique color on a labeled texture. The proxy ground-truth heatmap is used to train a deep keypoint prediction network, which can be used in online inference. The proposed approach to camera pose prediction show significant improvements when compared with state-of-the-art methods. Our approach to camera pose prediction is used to infer 3D objects from 2D image frames of video sequences online. To train the reconstruction model, it receives only a silhouette mask from a single frame of a video sequence in every training step and a category-specific mean object shape. We conducted experiments using three different datasets representing the bird category: the CUB [51] image dataset, YouTubeVos and the Davis video datasets. The network is trained on the CUB dataset and tested on all three datasets. The online experiments are demonstrated on YouTubeVos and Davis [56] video sequences using a network trained on the CUB training set.

Ludvig Dillén, Per-Erik Forssén, Johan Edstedt

We present FACT, a method for predicting alignment quality (i.e., registration error) of registered lidar point cloud pairs. This is useful e.g. for quality assurance of large, automatically registered 3D models. FACT extracts local features from a registered pair and processes them with a point transformer-based network to predict a misalignment class. We generalize prior work that study binary alignment classification of registration errors, by recasting it as multinomial misalignment classification. To achieve this, we introduce a custom regression-by-classification loss function that combines the cross-entropy and Wasserstein losses, and demonstrate that it outperforms both direct regression and prior binary classification. FACT successfully classifies point-cloud pairs registered with both the classical ICP and GeoTransformer, while other choices, such as standard point-cloud-quality metrics and registration residuals are shown to be poor choices for predicting misalignment. On a synthetically perturbed point-cloud task introduced by the CorAl method, we show that FACT achieves substantially better performance than CorAl. Finally, we demonstrate how FACT can assist experts in correcting misaligned point-cloud maps. Our code is available at https://github.com/LudvigDillen/FACT_for_PCMC.

Viktor Bergkvist, Felix Rydell, Per-Erik Forssén et al.

Event cameras offer distinct advantages over conventional frame-based sensors, including microsecond-level temporal resolution, high dynamic range, and low bandwidth. In this paper, we predict per-pixel depth distributions from monocular event streams using deep neural networks. We estimate uncertainty using Gaussian, log-normal, and evidential learning frameworks. We compare six event representations: spatio-temporal voxel grids with 1, 5, 10, and 20 temporal bins, the Compact Spatio-Temporal Representation (CSTR), and Time-Ordered Recent Event (TORE) volumes. Our U-Net-based models are trained on synthetic data and then fine-tuned on real sequences. We evaluate performance using absolute relative error, root mean squared error, and the area under the sparsification error. Quantitative results show that the representations perform similarly, while 10 bin log-normal and 5 bin evidential learning perform best across metrics. Our experiments demonstrate that uncertainty estimation can be successfully integrated into event-based monocular depth estimation, and be used to indicate pixels with reliable depth.

Yushan Zhang, Johan Edstedt, Bastian Wandt et al.

We tackle the task of scene flow estimation from point clouds. Given a source and a target point cloud, the objective is to estimate a translation from each point in the source point cloud to the target, resulting in a 3D motion vector field. Previous dominant scene flow estimation methods require complicated coarse-to-fine or recurrent architectures as a multi-stage refinement. In contrast, we propose a significantly simpler single-scale one-shot global matching to address the problem. Our key finding is that reliable feature similarity between point pairs is essential and sufficient to estimate accurate scene flow. We thus propose to decompose the feature extraction step via a hybrid local-global-cross transformer architecture which is crucial to accurate and robust feature representations. Extensive experiments show that the proposed Global Matching Scene Flow (GMSF) sets a new state-of-the-art on multiple scene flow estimation benchmarks. On FlyingThings3D, with the presence of occlusion points, GMSF reduces the outlier percentage from the previous best performance of 27.4% to 5.6%. On KITTI Scene Flow, without any fine-tuning, our proposed method shows state-of-the-art performance. On the Waymo-Open dataset, the proposed method outperforms previous methods by a large margin. The code is available at https://github.com/ZhangYushan3/GMSF.

Felix Järemo Lawin, Per-Erik Forssén

Probabilistic methods for point set registration have interesting theoretical properties, such as linear complexity in the number of used points, and they easily generalize to joint registration of multiple point sets. In this work, we improve their recognition performance to match state of the art. This is done by incorporating learned features, by adding a von Mises-Fisher feature model in each mixture component, and by using learned attention weights. We learn these jointly using a registration loss learning strategy (RLL) that directly uses the registration error as a loss, by back-propagating through the registration iterations. This is possible as the probabilistic registration is fully differentiable, and the result is a learning framework that is truly end-to-end. We perform extensive experiments on the 3DMatch and Kitti datasets. The experiments demonstrate that our approach benefits significantly from the integration of the learned features and our learning strategy, outperforming the state-of-the-art on Kitti. Code is available at https://github.com/felja633/RLLReg.

Felix Järemo Lawin, Martin Danelljan, Fahad Shahbaz Khan et al.

Probabilistic methods for point set registration have demonstrated competitive results in recent years. These techniques estimate a probability distribution model of the point clouds. While such a representation has shown promise, it is highly sensitive to variations in the density of 3D points. This fundamental problem is primarily caused by changes in the sensor location across point sets. We revisit the foundations of the probabilistic registration paradigm. Contrary to previous works, we model the underlying structure of the scene as a latent probability distribution, and thereby induce invariance to point set density changes. Both the probabilistic model of the scene and the registration parameters are inferred by minimizing the Kullback-Leibler divergence in an Expectation Maximization based framework. Our density-adaptive registration successfully handles severe density variations commonly encountered in terrestrial Lidar applications. We perform extensive experiments on several challenging real-world Lidar datasets. The results demonstrate that our approach outperforms state-of-the-art probabilistic methods for multi-view registration, without the need of re-sampling. Code is available at https://github.com/felja633/DARE.

Felix Järemo Lawin, Per-Erik Forssén, Hannes Ovrén

In this paper we introduce an efficient method to unwrap multi-frequency phase estimates for time-of-flight ranging. The algorithm generates multiple depth hypotheses and uses a spatial kernel density estimate (KDE) to rank them. The confidence produced by the KDE is also an effective means to detect outliers. We also introduce a new closed-form expression for phase noise prediction, that better fits real data. The method is applied to depth decoding for the Kinect v2 sensor, and compared to the Microsoft Kinect SDK and to the open source driver libfreenect2. The intended Kinect v2 use case is scenes with less than 8m range, and for such cases we observe consistent improvements, while maintaining real-time performance. When extending the depth range to the maximal value of 8.75m, we get about 52% more valid measurements than libfreenect2. The effect is that the sensor can now be used in large depth scenes, where it was previously not a good choice. Code and supplementary material are available at http://www.cvl.isy.liu.se/research/datasets/kinect2-dataset.

Emil Brissman, Per-Erik Forssén, Johan Edstedt

A challenge in image based metrology and forensics is intrinsic camera calibration when the used camera is unavailable. The unavailability raises two questions. The first question is how to find the projection model that describes the camera, and the second is to detect incorrect models. In this work, we use off-the-shelf extended PnP-methods to find the model from 2D-3D correspondences, and propose a method for model validation. The most common strategy for evaluating a projection model is comparing different models' residual variances - however, this naive strategy cannot distinguish whether the projection model is potentially underfitted or overfitted. To this end, we model the residual errors for each correspondence, individually scale all residuals using a predicted variance and test if the new residuals are drawn from a standard normal distribution. We demonstrate the effectiveness of our proposed validation in experiments on synthetic data, simulating 2D detection and Lidar measurements. Additionally, we provide experiments using data from an actual scene and compare non-camera access and camera access calibrations. Last, we use our method to validate annotations in MegaDepth.

Johannes Hägerlind, Bao-Long Tran, Urs Waldmann et al.

Estimating camera intrinsics and extrinsics is a fundamental problem in computer vision, and while advances in structure-from-motion (SfM) have improved accuracy and robustness, open challenges remain. In this paper, we introduce a robust method for pose estimation and calibration. We consider a set of rigid cameras, each observing the scene from a different perspective, which is a typical camera setup in animal behavior studies and forensic analysis of surveillance footage. Specifically, we analyse the individual components in a structure-from-motion (SfM) pipeline, and identify design choices that improve accuracy. Our main contributions are: (1) we investigate how to best subsample the predicted correspondences from a dense matcher to leverage them in the estimation process. (2) We investigate selection criteria for how to add the views incrementally. In a rigorous quantitative evaluation, we show the effectiveness of our changes, especially for cameras with strong radial distortion (79.9% ours vs. 40.4 vanilla VGGT). Finally, we demonstrate our correspondence subsampling in a global SfM setting where we initialize the poses using VGGT. The proposed pipeline generalizes across a wide range of camera setups, and could thus become a useful tool for animal behavior and forensic analysis.

Simon Kristoffersson Lind, Ziliang Xiong, Per-Erik Forssén et al.

When deploying deep neural networks on robots or other physical systems, the learned model should reliably quantify predictive uncertainty. A reliable uncertainty allows downstream modules to reason about the safety of its actions. In this work, we address metrics for evaluating such an uncertainty. Specifically, we focus on regression tasks, and investigate Area Under Sparsification Error (AUSE), Calibration Error, Spearman's Rank Correlation, and Negative Log-Likelihood (NLL). Using synthetic regression datasets, we look into how those metrics behave under four typical types of uncertainty, their stability regarding the size of the test set, and reveal their strengths and weaknesses. Our results indicate that Calibration Error is the most stable and interpretable metric, but AUSE and NLL also have their respective use cases. We discourage the usage of Spearman's Rank Correlation for evaluating uncertainties and recommend replacing it with AUSE.

Shipeng Liu, Ziliang Xiong, Bastian Wandt et al.

Human Pose Estimation (HPE) is increasingly important for applications like virtual reality and motion analysis, yet current methods struggle with balancing accuracy, computational efficiency, and reliable uncertainty quantification (UQ). Traditional regression-based methods assume fixed distributions, which might lead to poor UQ. Heatmap-based methods effectively model the output distribution using likelihood heatmaps, however, they demand significant resources. To address this, we propose Continuous Flow Residual Estimation (CFRE), an integration of Continuous Normalizing Flows (CNFs) into regression-based models, which allows for dynamic distribution adaptation. Through extensive experiments, we show that CFRE leads to better accuracy and uncertainty quantification with retained computational efficiency on both 2D and 3D human pose estimation tasks.

Shipeng Liu, Ziliang Xiong, Khac-Hoang Ngo et al.

Point cloud registration (PCR) is crucial for many downstream tasks, such as simultaneous localization and mapping (SLAM) and object tracking. This makes detecting and quantifying registration misalignment, i.e., PCR quality validation, an important task. All existing methods treat validation as a classification task, aiming to assign the PCR quality to a few classes. In this work, we instead use regression for PCR validation, allowing for a more fine-grained quantification of the registration quality. We also extend previously used misalignment-related features by using multiscale extraction and attention-based aggregation. This leads to accurate and robust registration error estimation on diverse datasets, especially for point clouds with heterogeneous spatial densities. Furthermore, when used to guide a mapping downstream task, our method significantly improves the mapping quality for a given amount of re-registered frames, compared to the state-of-the-art classification-based method.

Hannes Ovrén, Per-Erik Forssén

This paper revisits the problem of continuous-time structure from motion, and introduces a number of extensions that improve convergence and efficiency. The formulation with a $\mathcal{C}^2$-continuous spline for the trajectory naturally incorporates inertial measurements, as derivatives of the sought trajectory. We analyse the behaviour of split interpolation on $\mathbb{SO}(3)$ and on $\mathbb{R}^3$, and a joint interpolation on $\mathbb{SE}(3)$, and show that the latter implicitly couples the direction of translation and rotation. Such an assumption can make good sense for a camera mounted on a robot arm, but not for hand-held or body-mounted cameras. Our experiments show that split interpolation on $\mathbb{SO}(3)$ and on $\mathbb{R}^3$ is preferable over $\mathbb{SE}(3)$ interpolation in all tested cases. Finally, we investigate the problem of landmark reprojection on rolling shutter cameras, and show that the tested reprojection methods give similar quality, while their computational load varies by a factor of 2.

Hannes Ovrén, Per-Erik Forssén

In this paper we derive and test a probability-based weighting that can balance residuals of different types in spline fitting. In contrast to previous formulations, the proposed spline error weighting scheme also incorporates a prediction of the approximation error of the spline fit. We demonstrate the effectiveness of the prediction in a synthetic experiment, and apply it to visual-inertial fusion on rolling shutter cameras. This results in a method that can estimate 3D structure with metric scale on generic first-person videos. We also propose a quality measure for spline fitting, that can be used to automatically select the knot spacing. Experiments verify that the obtained trajectory quality corresponds well with the requested quality. Finally, by linearly scaling the weights, we show that the proposed spline error weighting minimizes the estimation errors on real sequences, in terms of scale and end-point errors.