Supplemental Material for Learning Online Smooth Predictors for Realtime Camera Planning using Recurrent Decision Trees
نویسندگان
چکیده
The autoregresor fπ(y−1, . . . , y−τ ) is typically selected from a class of autoregressors F . In our experiments, we use regularized linear autoregressors as F . Consider a generic learning policy π̂ with rolled-out trajectory Ŷ = {ŷt}t=1 corresponding to the input sequence X = {xt}t=1. We form the state sequence S = {st}t=1 = {[xt, . . . , xt−τ , ŷt−1, . . . , ŷt−τ ]}t=1. We approximate the smoothness of the curve Ŷ by a linear autoregressor
منابع مشابه
Learning Online Smooth Predictors for Realtime Camera Planning
We study the problem of online prediction for realtime camera planning, where the goal is to predict smooth trajectories that correctly track and frame objects of interest (e.g., players in a basketball game). The conventional approach for training predictors does not directly consider temporal consistency, and often produces undesirable jitter. Although post-hoc smoothing (e.g., via a Kalman f...
متن کاملEconomic analysis of the optimal level of supplemental irrigation for rain-fed figs
ABSTRACT- The aim of this study is economic analysis of supplemental irrigation (SI) for rain-fed fig trees in the south of Iran, with and without micro-catchment water harvesting systems (MCWHS). Under no MCWHS, by decreasing about 55 % of applied water, the fig yield decreased about 28% and net income increased twice compared with the maximum yield condition. In general, the ...
متن کاملLearning Smooth, Human-Like Turntaking in Realtime Dialogue
Giving synthetic agents human-like realtime turntaking skills is a challenging task. Attempts have been made to manually construct such skills, with systematic categorization of silences, prosody and other candidate turn-giving signals, and to use analysis of corpora to produce static decision trees for this purpose. However, for general-purpose turntaking skills which vary between individuals ...
متن کاملA model-based approximate λ-policy iteration approach to online evasive path planning and the video game Ms. Pac-Man
This paper presents a model-based approximate λ-policy iteration approach using temporal differences for optimizing paths online for a pursuit-evasion problem, where an agent must visit several target positions within a region of interest while simultaneously avoiding one or more actively pursuing adversaries. This method is relevant to applications, such as robotic path planning, mobile-sensor...
متن کاملInteractive Learning for Sequential Decisions and Predictions
Sequential prediction problems arise commonly in many areas of robotics and information processing: e.g., predicting a sequence of actions over time to achieve a goal in a control task, interpreting an image through a sequence of local image patch classifications, or translating speech to text through an iterative decoding procedure. Learning predictors that can reliably perform such sequential...
متن کامل