arXiv subjects

David Gustafsson

Publications and source records attributed to David Gustafsson.

8 recordsLinked to original sources

arXiv2026

Residual Kalman Dynamics for Event-Based UAV Forecasting

We study short- and mid-horizon UAV bounding-box forecasting on the FRED event-camera dataset. We use a constant-velocity Kalman filter over a full center-size box state as a strong physical baseline, and train a residual model to predict acceleration-like corrections from recent box history, filtered state features, and local event representations. This simple residual formulation consistently improves over the Kalman baseline, with event-conditioned models giving the strongest results among the evaluated methods. We further show that part of the residual target is predictable from anchor position and velocity alone, indicating that canonical FRED results can reflect both visual evidence and dataset-specific motion priors. To analyze this effect, we introduce decorrelated subsets as a diagnostic stress test, showing that event-conditioned residual models retain useful predictive signal even when measured position- and velocity-based shortcuts are weakened.

cs.CV↗

arXiv2026

Incremental Recommendation via Causal Models

Recommendation impressions are a finite resource, hence delivering a recommendation to a user who would discover the content organically yields no incremental value and displaces other recommendations that could. We address this by extending an existing production recommendation model to a causal architecture using holdback data that is already collected as part of routine experimentation infrastructure, requiring no new data collection. A central challenge is that attribution windows differ between treated and holdback observations: treated users are attributed a stream within a short direct-response window, while holdback users are attributed organic streams over a multi-day window. This mismatch makes naive treatment-effect subtraction invalid. We resolve this with a dual-threshold targeting policy that delivers a recommendation only when the probability of a treated stream is high and the probability of organic stream is low. In a production-scale A/B test on millions of Spotify users, this policy reduces recommendation impressions by 7% with no statistically significant reduction in overall recommended content consumption. We further show that joint training with holdback data improves calibration of the treated head relative to the production baseline, and argue this can be taken as evidence that causal models learn more generalisable representations than models trained on observational data alone.

stat.ML↗

arXiv2026

Neuromorphic Monocular Depth Estimation with Uncertainty Modeling

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.

cs.CV↗

arXiv2025

Measuring Uncertainty Calibration

We make two contributions to the problem of estimating the $L_1$ calibration error of a binary classifier from a finite dataset. First, we provide an upper bound for any classifier where the calibration function has bounded variation. Second, we provide a method of modifying any classifier so that its calibration error can be upper bounded efficiently without significantly impacting classifier performance and without any restrictive assumptions. All our results are non-asymptotic and distribution-free. We conclude by providing advice on how to measure calibration error in practice. Our methods yield practical procedures that can be run on real-world datasets with modest overhead.

cs.LG↗

arXiv2025

Drone Detection Using a Low-Power Neuromorphic Virtual Tripwire

Small drones are an increasing threat to both military personnel and civilian infrastructure, making early and automated detection crucial. In this work we develop a system that uses spiking neural networks and neuromorphic cameras (event cameras) to detect drones. The detection model is deployed on a neuromorphic chip making this a fully neuromorphic system. Multiple detection units can be deployed to create a virtual tripwire which detects when and where drones enter a restricted zone. We show that our neuromorphic solution is several orders of magnitude more energy efficient than a reference solution deployed on an edge GPU, allowing the system to run for over a year on battery power. We investigate how synthetically generated data can be used for training, and show that our model most likely relies on the shape of the drone rather than the temporal characteristics of its propellers. The small size and low power consumption allows easy deployment in contested areas or locations that lack power infrastructure.

cs.CV↗

arXiv2019

Unpaired Thermal to Visible Spectrum Transfer using Adversarial Training

Thermal Infrared (TIR) cameras are gaining popularity in many computer vision applications due to their ability to operate under low-light conditions. Images produced by TIR cameras are usually difficult for humans to perceive visually, which limits their usability. Several methods in the literature were proposed to address this problem by transforming TIR images into realistic visible spectrum (VIS) images. However, existing TIR-VIS datasets suffer from imperfect alignment between TIR-VIS image pairs which degrades the performance of supervised methods. We tackle this problem by learning this transformation using an unsupervised Generative Adversarial Network (GAN) which trains on unpaired TIR and VIS images. When trained and evaluated on KAIST-MS dataset, our proposed methods was shown to produce significantly more realistic and sharp VIS images than the existing state-of-the-art supervised methods. In addition, our proposed method was shown to generalize very well when evaluated on a new dataset of new environments.

cs.CV↗

arXiv2011

Soft nanostructuring of YBCO Josephson Junctions by phase separation

We have developed a new method to fabricate biepitaxial YBa2Cu3O(7-x) (YBCO) Josephson junctions at the nanoscale, allowing junctions widths down to 100 nm and simultaneously avoiding the typical damage in grain boundary interfaces due to conventional patterning procedures. By using the competition between the superconducting YBCO and the insulating Y2BaCuO5 phases during film growth, we formed nanometer sized grain boundary junctions in the insulating Y2BaCuO5 matrix as confirmed by high resolution transmission electron microscopy. Electrical transport measurements give clear indications that we are close to probing the intrinsic properties of the grain boundaries.

cond-mat.supr-con↗

arXiv2011

Noise properties of nanoscale YBCO Josephson junctions

We present electric noise measurements of nanoscale biepitaxial YBa2Cu3O(7-x) (YBCO) Josephson junctions fabricated by two different lithographic methods. The first (conventional) technique defines the junctions directly by ion milling etching through an amorphous carbon mask. The second (soft patterning) method makes use of the phase competition between the superconducting YBCO (Y123) and the insulating Y2BaCuO5 (Y211) phase at the grain boundary interface on MgO (110) substrates. The voltage noise properties of the two methods are compared in this study. For all junctions (having a thickness of 100 nm and widths of 250-500 nm) we see a significant amount of individual charge traps. We have extracted an approximate value for the effective area of the charge traps from the noise data. From the noise measurements we infer that the soft patterned junctions with a grain boundary (GB) interface manifesting a large c-axis tunneling component have a uniform barrier and a SIS like behavior. The noise properties of soft patterned junctions having a GB interface dominated by transport parallel to the ab-planes are in accordance with a resonant tunneling barrier model. The conventionally patterned junctions, instead, have suppressed superconducting transport channels with an area much less than the nominal junction area. These findings are important for the implementation of nanosized Josephson junctions in quantum circuits.

cond-mat.supr-con↗