A comprehensive handball dynamics dataset for game situation classification.

This article presents a comprehensive dataset of labeled game situations obtained from multiple professional handball matches, which corresponds to the research paper entitled "PlayNet: Real-time Handball Play Classification with Kalman Embeddings and Neural Networks" [1]. The dataset encompasses approximately 11 hours of footage from five handball games played in two different arenas, resulting in around 1 million data frames. Each frame has been meticulously labeled using seven distinct game situation classes (left and right attacks, left and right transitions, left and right penalties, and timeouts). Notably, the dataset does not contain video frames, but provides a synthetic normalized representation of each frame. This representation includes information about player, referee, and ball positions, as well as player and referee velocities, for every labeled game situation. We obtained said details automatically by using an object detector to infer the positions of players, referees, and the ball in each frame. After tracking the detected agent positions across frames, the extracted coordinates underwent normalization through a "bird's eye" perspective transform, ensuring that the data remained unaffected by variations in camera configurations across different arenas. Finally, a Kalman filter was applied to improve the robustness of player positions and derive their velocities. The labeling process was performed by domain experts employing a custom system designed to annotate game situations, considering the play type and its contextual setting. In conclusion, researchers can utilize this dataset for several purposes: game analysis, automated broadcasting, or game summarization. Furthermore, this dataset can contribute to a broader understanding of the relationship between player dynamics and game situations, shedding light on the level of granularity required for accurately classifying them.

Real-Time Denoising of Volumetric Path Tracing for Direct Volume Rendering.

Direct volume rendering (DVR) using volumetric path tracing (VPT) is a scientific visualization technique that simulates light transport with objects' matter using physically-based lighting models. Monte Carlo (MC) path tracing is often used with surface models, yet its application for volumetric models is difficult due to the complexity of integrating MC light-paths in volumetric media with none or smooth material boundaries. Moreover, auxiliary geometry-buffers (G-buffers) produced for volumes are typically very noisy, failing to guide image denoisers relying on that information to preserve image details. This makes existing real-time denoisers, which take noise-free G-buffers as their input, less effective when denoising VPT images. We propose the necessary modifications to an image-based denoiser previously used when rendering surface models, and demonstrate effective denoising of VPT images. In particular, our denoising exploits temporal coherence between frames, without relying on noise-free G-buffers, which has been a common assumption of existing denoisers for surface-models. Our technique preserves high-frequency details through a weighted recursive least squares that handles heterogeneous noise for volumetric models. We show for various real data sets that our method improves the visual fidelity and temporal stability of VPT during classic DVR operations such as camera movements, modifications of the light sources, and editions to the volume transfer function.

Differences in pollination success between local and foreign flower color phenotypes: a translocation experiment with Gentiana lutea (Gentianaceae).

BACKGROUND: The adaptive maintenance of flower color variation is frequently attributed to pollinators partly because they preferentially visit certain flower phenotypes. We tested whether Gentiana lutea-which shows a flower color variation (from orange to yellow) in the Cantabrian Mountains range (north of Spain)-is locally adapted to the pollinator community. METHODS: We transplanted orange-flowering individuals to a population with yellow-flowering individuals and vice versa, in order to assess whether there is a pollination advantage in the local morph by comparing its visitation rate with the foreign morph. RESULTS: Our reciprocal transplant experiment did not show clear local morph advantage in overall visitation rate: local orange flowers received more visits than foreign yellow flowers in the orange population, while both local and foreign flowers received the same visits in the yellow population; thus, there is no evidence of local adaptation in Gentiana lutea to the pollinator assemblage. However, some floral visitor groups (such as Bombus pratorum, B. soroensis ancaricus and B. lapidarius decipiens) consistently preferred the local morph to the foreign morph whereas others (such as Bombus terrestris) consistently preferred the foreign morph. DISCUSSION: We concluded that there is no evidence of local adaptation to the pollinator community in each of the two G. lutea populations studied. The consequences for local adaptation to pollinator on G. lutea flower color would depend on the variation along the Cantabrian Mountains range in morph frequency and pollinator community composition.

Flower color preferences of insects and livestock: effects on Gentiana lutea reproductive success.

Angiosperms diversification was primarily driven by pollinator agents, but non-pollinator agents also promoted floral evolution. Gentiana lutea shows pollinator driven flower color variation in NW Spain. We test whether insect herbivores and livestock, which frequently feed in G.lutea, play a role in G. lutea flower color variation, by answering the following questions: (i) Do insect herbivores and grazing livestock show flower color preferences when feeding on G. lutea? (ii) Do mutualists (pollinators) and antagonists (seed predators, insect herbivores and livestock) jointly affect G. lutea reproductive success? Insect herbivores fed more often on yellow flowering individuals but they did not affect seed production, whereas livestock affected seed production but did not show clear color preferences. Our data indicate that flower color variation of G. lutea is not affected by insect herbivores or grazing livestock.

Is there a hybridization barrier between Gentiana lutea color morphs?

In Gentiana lutea two varieties are described: G. lutea var. aurantiaca with orange corolla colors and G. lutea var. lutea with yellow corolla colors. Both color varieties co-occur in NW Spain, and pollinators select flower color in this species. It is not known whether a hybridization barrier exists between these G. lutea color varieties. We aim to test the compatibility between flower color varieties in G. lutea and its dependence on pollen vectors. Within a sympatric population containing both flower color morphs, we analyzed differences in reproductive success (number, weight, viability and germinability of seeds) depending on fertilization treatments (autogamy and xenogamy within variety and among varieties). We found a 93% reduction in number of seeds and a 37% reduction in seed weight respectively of autogamy treatments compared to xenogamy crossings. Additionally, reproductive success is higher within color varieties than among varieties, due to a 45% seed viability reduction on hybrids from different varieties. Our results show that G. lutea reproductive success is strongly dependent on pollinators and that a partial hybridization barrier exists between G. lutea varieties.

Selective Pressures Explain Differences in Flower Color among Gentiana lutea Populations.

Flower color variation among plant populations might reflect adaptation to local conditions such as the interacting animal community. In the northwest Iberian Peninsula, flower color of Gentiana lutea varies longitudinally among populations, ranging from orange to yellow. We explored whether flower color is locally adapted and the role of pollinators and seed predators as agents of selection by analyzing the influence of flower color on (i) pollinator visitation rate and (ii) escape from seed predation and (iii) by testing whether differences in pollinator communities correlate with flower color variation across populations. Finally, (iv) we investigated whether variation in selective pressures explains flower color variation among 12 G. lutea populations. Flower color influenced pollinator visits and differences in flower color among populations were related to variation in pollinator communities. Selective pressures on flower color vary among populations and explain part of flower color differences among populations of G. lutea. We conclude that flower color in G. lutea is locally adapted and that pollinators play a role in this adaptation.

Selective pressure along a latitudinal gradient affects subindividual variation in plants.

Individual plants produce repeated structures such as leaves, flowers or fruits, which, although belonging to the same genotype, are not phenotypically identical. Such subindividual variation reflects the potential of individual genotypes to vary with micro-environmental conditions. Furthermore, variation in organ traits imposes costs to foraging animals such as time, energy and increased predation risk. Therefore, animals that interact with plants may respond to this variation and affect plant fitness. Thus, phenotypic variation within an individual plant could be, in part, an adaptive trait. Here we investigated this idea and we found that subindividual variation of fruit size of Crataegus monogyna, in different populations throughout the latitudinal gradient in Europe, was explained at some extent by the selective pressures exerted by seed-dispersing birds. These findings support the hypothesis that within-individual variation in plants is an adaptive trait selected by interacting animals which may have important implications for plant evolution.

Functional diversity among seed dispersal kernels generated by carnivorous mammals.

1. Knowledge of the spatial scale of the dispersal service provided by important seed dispersers (i.e. common and/or keystone species) is essential to our understanding of their role on plant ecology, ecosystem functioning and, ultimately, biodiversity conservation. 2. Carnivores are the main mammalian frugivores and seed dispersers in temperate climate regions. However, information on the seed dispersal distances they generate is still very limited. We focused on two common temperate carnivores differing in body size and spatial ecology - red fox (Vulpes vulpes) and European pine marten (Martes martes) - for evaluating possible functional diversity in their seed dispersal kernels. 3. We measured dispersal distances using colour-coded seed mimics embedded in experimental fruits that were offered to the carnivores in feeding stations (simulating source trees). The exclusive colour code of each simulated tree allowed us to assign the exact origin of seed mimics found later in carnivore faeces. We further designed an explicit sampling strategy aiming to detect the longest dispersal events; as far we know, the most robust sampling scheme followed for tracking carnivore-dispersed seeds. 4. We found a marked functional heterogeneity among both species in their seed dispersal kernels according to their home range size: multimodality and long-distance dispersal in the case of the fox and unimodality and short-distance dispersal in the case of the marten (maximum distances = 2846 and 1233 m, respectively). As a consequence, emergent kernels at the guild level (overall and in two different years) were highly dependent on the relative contribution of each carnivore species. 5. Our results provide the first empirical evidence of functional diversity among seed dispersal kernels generated by carnivorous mammals. Moreover, they illustrate for the first time how seed dispersal kernels strongly depend on the relative contribution of different disperser species, thus on the composition of local disperser assemblages. These findings provide a key starting point for understanding and modelling plant population processes that include mammal-mediated seed dispersal, such as connectivity, range expansion and colonization.

Interactive multiscale tensor reconstruction for multiresolution volume visualization.

Large scale and structurally complex volume datasets from high-resolution 3D imaging devices or computational simulations pose a number of technical challenges for interactive visual analysis. In this paper, we present the first integration of a multiscale volume representation based on tensor approximation within a GPU-accelerated out-of-core multiresolution rendering framework. Specific contributions include (a) a hierarchical brick-tensor decomposition approach for pre-processing large volume data, (b) a GPU accelerated tensor reconstruction implementation exploiting CUDA capabilities, and (c) an effective tensor-specific quantization strategy for reducing data transfer bandwidth and out-of-core memory footprint. Our multiscale representation allows for the extraction, analysis and display of structural features at variable spatial scales, while adaptive level-of-detail rendering methods make it possible to interactively explore large datasets within a constrained memory footprint. The quality and performance of our prototype system is evaluated on large structurally complex datasets, including gigabyte-sized micro-tomographic volumes.