Creating group choreography from music is crucial in cultural entertainment and virtual reality, with a focus on generating harmonious movements. Despite growing interest, recent approaches often struggle with two major challenges: multi-dancer collisions and single-dancer foot sliding. To address these challenges, we propose a Trajectory-Controllable Diffusion (TCDiff) framework, which leverages non-overlapping trajectories to ensure coherent and aesthetically pleasing dance movements. To mitigate collisions, we introduce a Dance-Trajectory Navigator that generates collision-free trajectories for multiple dancers, utilizing a distance-consistency loss to maintain optimal spacing. Furthermore, to reduce foot sliding, we present a footwork adaptor that adjusts trajectory displacement between frames, supported by a relative forward-kinematic loss to further reinforce the correlation between movements and trajectories. Experiments demonstrate our method's superiority.

The inherent stochasticity of the agents' behavior presents a challenge to trajectory prediction models, which are required to generate multiple plausible future trajectories. Recently, diffusion models have been applied to implement multimodal trajectory prediction. Existing approaches typically employ a standard diffusion process, denoising from a sample drawn from a Gaussian distribution. However, we identify that most agents exhibit an obvious movement trend, rendering many initial denoising steps redundant-primarily transitioning from pure noise to an initial coarse trajectory. To conquer this challenge, this paper innovatively proposes a diffusion refiner that can be used along with existing multi-agent trajectory prediction models to improve their performance. Specifically, we first leverage a baseline model for predicting the coarse future trajectory. Then, the diffusion model is applied as a refiner to reduce the prediction error. Moreover, our method is naturally plug-and-play, allowing convenient integration with existing models. To achieve this, we improve the traditional diffusion process to not only converge towards noise but also the coarse predictions from the baseline model. In such a case, standard step-skipping sampling techniques is inapplicable and we further propose an ordinary differential equation (ODE)-based fast sampling method. Extensive experiments with selected baseline models demonstrate the effectiveness of our approach.

Trajectory prediction, which aims to predict the future positions of all agents in a crowd scene, given their past trajectories, plays a vital role in improving the safety of autonomous driving vehicles. For heterogeneous agents, it is imperative to account for the gap in feature distribution differences between agents in different categories. Besides, exploring the reference relationship between the future motions of agents is crucial yet overlooked in previous trajectory prediction methods. To tackle these challenges, we propose an edge-enhanced heterogeneous graph Transformer with priority-based feature aggregation for multi-modal trajectory prediction. Specifically, a new edge-enhanced heterogeneous interaction module that carries relative position information via edges is proposed to explore the complex interaction among agents. Additionally, we propose the concept of priority during the decoding phase and the corresponding measuring method, based on which a priority-based feature aggregation module is presented to enable referencing between agents, allowing for a more reasonable trajectory generation process. Additionally, we design an effective feature fusion method based on state refinement LSTM so that temporal and social features can be well integrated while accounting for their roles in trajectory prediction. Extensive experimental results on public datasets demonstrate that our approach outperforms the state-of-the-art baseline methods, confirming the effectiveness of our proposed method. The source code of our EPHGT model will be publicly released athttps://github.com/xbchen82/EPHGT.