Recognition of handwritten mathematical expressions allows to transfer scientific notes into their digital form. It facilitates the sharing, searching, and preservation of scientific information. We introduce MathWriting, the largest online handwritten mathematical expression dataset to date. It consists of 230k human-written samples and an additional 400k synthetic ones. This dataset can also be used in its rendered form for offline HME recognition. One MathWriting sample consists of a formula written on a touch screen and a corresponding LaTeX expression. We also provide a normalized version of LaTeX expression to simplify the recognition task and enhance the result quality. We provide baseline performance of standard models like OCR and CTC Transformer as well as Vision-Language Models like PaLI on the dataset. The dataset together with an example colab is accessible on Github.

Ink is a powerful medium for note-taking and creativity tasks. Multi-touch devices and stylus input have enabled digital ink to be editable and searchable. To extend the capabilities of digital ink, we introduce Inkeraction, an interaction modality powered by ink recognition and synthesis. Inkeraction segments and classifies digital ink objects (e.g., handwriting and sketches), identifies relationships between them, and generates strokes in different writing styles. Inkeraction reshapes the design space for digital ink by enabling features that include: (1) assisting users to manipulate ink objects, (2) providing word-processor features such as spell checking, (3) automating repetitive writing tasks such as transcribing, and (4) bridging with generative models’ features such as brainstorming. Feedback from two user studies with a total of 22 participants demonstrated that Inkeraction supported writing activities by enabling participants to write faster with fewer steps and achieve better writing quality.

Identity Switching remains one of the main difficulties Multiple Object Tracking (MOT) algorithms have to deal with. Many state-of-the-art approaches now use sequence models to solve this problem but their training can be affected by biases that decrease their efficiency. In this paper, we introduce a new training procedure that confronts the algorithm to its own mistakes while explicitly attempting to minimize the number of switches, which results in better training. We propose an iterative scheme of building a rich training set and using it to learn a scoring function that is an explicit proxy for the target tracking metric. Whether using only simple geometric features or more sophisticated ones that also take appearance into account, our approach outperforms the state-of-the-art on several MOT benchmarks.

People detection methods are highly sensitive to occlusions between pedestrians, which are extremely frequent in many situations where cameras have to be mounted at a limited height. The reduction of camera prices allows for the generalization of static multi-camera set-ups. Using joint visual information from multiple synchronized cameras gives the opportunity to improve detection performance. In this paper, we present a new large-scale and high-resolution dataset. It has been captured with seven static cameras in a public open area, and unscripted dense groups of pedestrians standing and walking. Together with the camera frames, we provide an accurate joint (extrinsic and intrinsic) calibration, as well as 7 series of 400 annotated frames for detection at a rate of 2 frames per second. This results in over 40 000 bounding boxes delimiting every person present in the area of interest, for a total of more than 300 individuals. We provide a series of benchmark results using baseline algorithms published over the recent months for multi-view detection with deep neural networks, and trajectory estimation using a non-Markovian model.

Many state-of-the-art approaches to multi-object tracking rely on detecting them in each frame independently, grouping detections into short but reliable trajectory segments, and then further grouping them into full trajectories. This grouping typically relies on imposing local smoothness constraints but almost never on enforcing more global ones on the trajectories. In this paper, we propose a non-Markovian approach to imposing global consistency by using behavioral patterns to guide the tracking algorithm. When used in conjunction with state-of-the-art tracking algorithms, this further increases their already good performance on multiple challenging datasets. We show significant improvements both in supervised settings where ground truth is available and behavioral patterns can be learned from it, and in completely unsupervised settings.

Tracking the ball is critical for video-based analysis of team sports. However, it is difficult, especially in low-resolution images, due to the small size of the ball, its speed that creates motion blur, and its often being occluded by players. In this paper, we propose a generic and principled approach to modeling the interaction between the ball and the players while also imposing appropriate physical constraints on the ball's trajectory. We show that our approach, formulated in terms of a Mixed Integer Program, is more robust and more accurate than several state-of-the-art approaches on real-life volleyball, basketball, and soccer sequences.

We investigate how metrics that can be measured offline can be used to predict the online performance of recommender systems, thus avoiding costly A-B testing. In addition to accuracy metrics, we combine diversity, coverage, and serendipity metrics to create a new performance model. Using the model, we quantify the trade-off between different metrics and propose to use it to tune the parameters of recommender algorithms without the need for online testing. Another application for the model is a self-adjusting algorithm blend that optimizes a recommender's parameters over time. We evaluate our findings on data and experiments from news websites.

In this paper, we introduce a novel approach for incident ticket classification that aims at minimizing the manual labelling effort while achieving good-quality predictions. To accomplish this, we devise a two-stage technique that employs hierarchical clustering using a combination of graph clustering (community finding) and topic modelling as first stage, followed by either another round of hierarchical clustering or an active learning approach as second stage. We evaluate the performance of our method in terms of manual labelling effort, prediction quality and efficiency on three real-world datasets and demonstrate that classical approaches to text classification are not well suited for incident ticket texts.