We introduce the first data-driven multi-view 3D point tracker, designed to track arbitrary points in dynamic scenes using multiple camera views. Unlike existing monocular trackers, which struggle with depth ambiguities and occlusion, or prior multi-camera methods that require over 20 cameras and tedious per-sequence optimization, our feed-forward model directly predicts 3D correspondences using a practical number of cameras (e.g., four), enabling robust and accurate online tracking. Given known camera poses and either sensor-based or estimated multi-view depth, our tracker fuses multi-view features into a unified point cloud and applies k-nearest-neighbors correlation alongside a transformer-based update to reliably estimate long-range 3D correspondences, even under occlusion. We train on 5K synthetic multi-view Kubric sequences and evaluate on two real-world benchmarks: Panoptic Studio and DexYCB, achieving median trajectory errors of 3.1 cm and 2.0 cm, respectively. Our method generalizes well to diverse camera setups of 1-8 views with varying vantage points and video lengths of 24-150 frames. By releasing our tracker alongside training and evaluation datasets, we aim to set a new standard for multi-view 3D tracking research and provide a practical tool for real-world applications. Project page available at https://ethz-vlg.github.io/mvtracker.

Machine learning (ML) models require large amounts of data and many of the stored data is used to train ML models. However, the ML models learn insights about the data during their training and this raises privacy concerns of the individuals regarding personal data. These concerns led to the introduction of legislation focusing on the “right to be forgotten” and machine unlearning has emerged to address these concerns. Although machine unlearning studies focus on data privacy issues generally, machine unlearning is also used to fix the mistrained machine learning models as well. Mistraining may occur due to problems in the data such as mislabeling. Machine unlearning can solve this problem by discarding the information regarding the problematic data. In this study, the effects of machine unlearning on facial attribute classification are discovered. Experimental results on CelebA dataset show the effectiveness of machine unlearning methods. The code repository can be accessed at https://github.com/ituvisionlab/face-attribute-unlearning.