Accurate long-term trajectory prediction in complex scenes, where multiple agents (e.g., pedestrians or vehicles) interact with each other and the environment while attempting to accomplish diverse often unknown goals, is a challenging stochastic forecasting problem. In this work, we propose MUSEVAE, new probabilistic modeling framework based on cascade of Conditional VAEs, which tackles long-term, uncertain task using coarse-to-fine multi-factor architecture. its Macro stage, model learns...

We present SEAN 2.0, an open-source system designed to advance social navigation via the training and benchmarking of policies in varied contexts. A key limitation current research is that are often trained evaluated considering only a few contexts, which fragmented across prior work. Inspired by work psychology, we describe context based on situations, encompass robot task environmental factors, propose logic-based classifiers for five common examples. 2.0 allows experience these situations...

FSS (Few-shot segmentation) aims to segment a target class using small number of labeled images (support set). To extract information relevant the class, dominant approach in best performing methods removes background features support mask. We observe that this feature excision through limiting mask introduces an bottleneck several challenging cases, e.g., for targets and/or inaccurate boundaries. end, we present novel method (MSI), which maximizes support-set by exploiting two complementary...

Landmarks play a vital role in human wayfinding by providing the structure for mental spatial representations and indicating locations with which to orient. Less research effort has been allocated towards automated landmark identification indoor environments despite growing interest navigation scientific community. In this paper, we propose computational framework identify landmarks that is based on hierarchical multi-criteria decision model grounded theories of cognition information...

For transportation hubs, leveraging pedestrian flows for commercial activities presents an effective strategy funding maintenance and infrastructure improvements. However, this introduces new challenges, as consumer behaviors can disrupt flow efficiency. To optimize both retail potential efficiency, careful strategic planning in store layout facility dimensions was done by expert judgement due to the complexity dynamics areas of hubs. This paper attention-based movement model simulate these...

In this paper, we propose StoryPrint, an interactive visualization of creative storytelling that facilitates individual and comparative structural analyses. This method is intended for script-based media, which has suitable metadata. The pre-visualization process involves parsing the script into different metadata categories analyzing sentiment on a character scene basis. For each scene, setting, presence, prominence, emotion film are represented as StoryPrint. presented radial diagram...

Previous studies regarding the perception of emotions for embodied virtual agents have shown effectiveness using characters in conveying through interactions with humans. However, creating an autonomous conversational agent expressive behaviors presents two major challenges. The first challenge is difficulty synthesizing each modality that are as real human behaviors. second affects modeled independently, which makes it difficult to generate multimodal responses consistent across all...

The collaboration between creatives and domain architects is crucial for bringing virtual characters to life. Domain are technical experts who tasked with formally designing intelligent characters' knowledge, which a symbolic representation of knowledge that the character uses reason over its interactions other agents. In context this work, encompasses mental modeling character. Although creation interactive narratives requires substantial engineering expertise, it also necessary pick brains...

Floorplans often require considering numerous factors, from the layout size to cost, numeric attributes such as room sizes, and other intrinsic properties connectivity between visible regions. Representing these complex factors is challenging, but doing so in a representative efficient way can enable new modes of design exploration. Existing image graph-based approaches floorplans’ representation failed consider low-level space semantics, structural features, utilization with respect its...

The development of autonomous agents for wayfinding tasks has long maintained the usage naive, omniscient models navigation. simplicity these improves scalability crowd simulations, but limits utility such simulations to visualization general behaviors. This restricted scope does not allow observation more nuanced, individualized In this paper, we demonstrate a novel framework agent that rely on omniscience. Instead, each is equipped with memory architecture enables by maintaining cognitive...

In recent years, human trajectory prediction (HTP) has garnered attention in computer vision literature. Although this task much common with the longstanding of crowd simulation, there is little from simulation that been borrowed, especially terms evaluation protocols. The key difference between two tasks HTP concerned forecasting multiple steps at a time and capturing multimodality real trajectories. A majority models are trained on same few datasets, which feature small, transient...

For transportation hubs, leveraging pedestrian flows for commercial activities presents an effective strategy funding maintenance and infrastructure improvements. However, this introduces new challenges, as consumer behaviors can disrupt flow efficiency. To optimize both retail potential efficiency, careful strategic planning in store layout facility dimensions was done by expert judgement due to the complexity dynamics areas of hubs. This paper attention-based movement model simulate these...

Digital storytelling, essential in entertainment, education, and marketing, faces challenges production scalability flexibility. The StoryAgent framework, introduced this paper, utilizes Large Language Models generative tools to automate refine digital storytelling. Employing a top-down story drafting bottom-up asset generation approach, tackles key issues such as manual intervention, interactive scene orchestration, narrative consistency. This framework enables efficient of consistent...

Accurate prediction of human or vehicle trajectories with good diversity that captures their stochastic nature is an essential task for many applications. However, trajectory models produce unreasonable samples focus on improving accuracy while neglecting other key requirements, such as collision avoidance the surrounding environment. In this work, we propose TrajDiffuse, a planning-based method using novel guided conditional diffusion model. We form problem denoising impaint and design...

Improving layout design focusing on human wayfinding is a non-trivial task because quantifying perception of space during movement depends multiple spatial attributes, such as the arrangement walls and area visible space. Existing approaches in computer-aided have yet to leverage role architectural configuration guide route-choice behavior support design. This paper presents novel optimization framework grounded cognition empower designers bridge gap between research. To facilitate proposed...

Predicting the behavior of crowds in complex environments is a key requirement multitude application areas, including crowd and disaster management, architectural design, urban planning. Given crowd's immediate state, current approaches simulate movement to arrive at future state. However, most applications require ability predict hundreds possible simulation outcomes (e.g., under different environment situations) real-time rates, for which these are prohibitively expensive. In this paper,...

Human path-planning operates differently from deterministic AI-based algorithms due to the decay and distortion in a human's spatial memory lack of complete scene knowledge. Here, we present cognitive model that simulates human-like learning unfamiliar environments, supports systematic degradation memory, distorts recall during path-planning. We propose Dynamic Hierarchical Cognitive Graph (DHCG) representation encode environment structure by incorporating two critical biases exploration:...