Liarokapis [7] proposed an augmented reality sys-
tem for a guitar learner. The aim of this work is to
show the augmentation (e.g., the positions where the
learner should place their ¯ngers to play the chord) on
an electric guitar to guide the player as a self-learning
system.
Motokawa and Saito [8] built a system called On-
line Guitar Tracking that supports a guitarist using
augmented reality. This is done by showing a virtual
model of the ¯ngers on a stringed guitar as an aid to
learning to play the guitar.
In these above systems, they do not aim to rec-
ognize the chord which a player is playing. Their
systems cannot determine whether the player is play-
ing chord correctly or not. A similar research goal as
ours can be found in [9]. Their goal is to develop an
electric bass guitar system to evaluate if the user is
pressing his/her fore¯nger on the correct fret of the
bass guitar. However, their system has a constraint
that only one ¯nger is used for playing. Our aim is
to propose a system for helping the player to play the
ordinary guitar (6 strings) that uses all four ¯ngers
for playing.