Multiple Object Forecasting: Predicting Future Object Locations in Diverse Environments

Abstract

Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

A Two-Block RNN-Based Trajectory Prediction From Incomplete Trajectory

“…The common solution for dealing with incomplete observed trajectory is to ignore these cases from the dataset as outliers. The agent who disappears even in one frame in a sequence is excluded from the dataset to handle incomplete data [1], [2], [12]. However, in real-world situations, an intelligent system must be able to continuously predict the future trajectory of all agents.…”

“…The problem of trajectory prediction has received significant attention in recent years across various applications, Trajectory prediction Gupta et al [2] Trajectory prediction Styles et al [12] Trajectory prediction Yao et al [13] Traffic accident detection Malla et al [14] Trajectory prediction Lipton et al [30] Medical data analysis Kim et al [31] Medical data analysis Che et al [32] Medical data analysis Cao et al [33] General Tian et al [34] Traffic flow prediction Kim et al [35] Medical data analysis Luo et al [36] General Luo et al [37] General Ours Trajectory prediction such as self-driving vehicles, service robots, and advanced surveillance systems. A large body of research has addressed this problem.…”

“…Many approaches have been developed to address this issue (see Table 1). If the task involves multiple time series (e.g., for trajectory prediction), a simple approach is to drop the missing data and perform the prediction using only the observed data [1], [2], [12]. However, this strategy leads to loss of information and may not work if the missing rate is high or when there are individual time series to process (e.g., in medical applications).…”

Multi-Camera Trajectory Forecasting with Trajectory Tensors

“…Other works also consider environmental constraints [21]. A comparatively small number of works consider visual features for trajectory forecasting, such as those extracted from human pose [9], [22] or optical flow [10] from non-nadir viewpoints. Models are often optimized using loss functions such as mean squared error (MSE) [9], [10], [23].…”

“…A comparatively small number of works consider visual features for trajectory forecasting, such as those extracted from human pose [9], [22] or optical flow [10] from non-nadir viewpoints. Models are often optimized using loss functions such as mean squared error (MSE) [9], [10], [23]. As the distribution of future trajectories is multimodal, using the MSE loss heavily penalizes reasonable but incorrect forecasts, such as turning right rather than left at an intersection.…”