Long Term Real Trajectory Reuse through Region Goal Satisfaction

Abstract

This paper is motivated by the objective of improving the realism of real-time simulated crowds by reducing short term collision avoidance through long term anticipation of pedestrian trajectories. For this aim, we choose to reuse outdoor pedestrian trajectories obtained with non-invasive means. This initial step is achieved by analyzing the recordings of multiple synchronized video cameras. In a second off-line stage, we fit as long as possible trajectory segments within predefined paths made of a succession of region goals. The concept of region goal is exploited to enforce the principle of "sufficient satisfaction": it allows the pedestrians to relax the prescribed trajectory to the traversal of successive region goals. However, even if a fitted trajectory is modified due to collision avoidance, we are still able to make long-term trajectory anticipation and distribute the collision avoidance shift over a long distance.

Within-crowd immersive evaluation of collision avoidance behaviors

“…Path variants (Figure 2 In the improved model, we increased the number of Real Trajectory Segments (RTS) by reducing rectification thresholds and by mirroring each segment. The thresholds of distance and speed local confidence functions [Ahn et al 2011] were decreased. In addition, RTSs are also classified by linearity, speed, goal distance, and the number of frames.…”

“…If we take a close look at this equation (see also Figure 6), the sum l k w k (ISV ijk ) represents the influence of the Character j to the Character i (ISVij). Four ISV ijk vectors from time tc to t h are illustrated in Figure 6, where the vectors' direction are generated by the equation of shift influence vector proposed in [Ahn et al 2011]. The weighting factor w k is the product of five different weights (w k1 w k2 w k3 w k4 w k5 ), each defined as follows:…”

“…When a collision is expected, we took the minimum magnitude possible by considering remaining time before collision. The magnitude of trajectory shift vector [Ahn et al 2011] is used for ISV magnitude generation. If no collisions are expected, we calculate a special ISV, which reduces the Shift Vector (SV) and helps the character to return to the original RTV line.…”

Presence Through Fullbodied Interactions Overview of research activities at the Immersive Interaction Group

“…The project is in the stage of trajectory reuse for synthesizing new crowds in virtual environments that are completely different from the original location [16]. Figure 7 illustrates the original capture location with reconstructed trajectories (left) and their reuse in a new environment (right).…”

A motion planning method for simulating a virtual crowd

“…Model proposed by Ahn et al (2011) presents a trajectory variant shift method by re-using and shifting real trajectories captured from video data to avoid collisions. Ahn et al (2012) gives the collision avoidance model by reducing short term collision avoidance through long term anticipation of pedestrian trajectories.…”