TY - GEN
T1 - Combining matching scores in identification model
AU - Tulyakov, Sergey
AU - Govindaraju, Venu
PY - 2005
Y1 - 2005
N2 - The paper discusses a problem of combining recognition scores for different classes produced by one recognizer during one recognition attempt. This problem arises in identification problems which we define as 1:N classification problems with big or variable N. By using artificial example we show that intuitive solution of making identification decision based solely on the best matching score is frequently suboptimal. Paper presents reasons for such behavior, and draws parallels with score normalization technique used in speaker identification. Two examples of real life applications illustrate the possible benefits of properly combining recognition scores.
AB - The paper discusses a problem of combining recognition scores for different classes produced by one recognizer during one recognition attempt. This problem arises in identification problems which we define as 1:N classification problems with big or variable N. By using artificial example we show that intuitive solution of making identification decision based solely on the best matching score is frequently suboptimal. Paper presents reasons for such behavior, and draws parallels with score normalization technique used in speaker identification. Two examples of real life applications illustrate the possible benefits of properly combining recognition scores.
UR - https://www.scopus.com/pages/publications/33845535601
U2 - 10.1109/ICDAR.2005.77
DO - 10.1109/ICDAR.2005.77
M3 - Conference contribution
AN - SCOPUS:33845535601
SN - 0769524206
SN - 9780769524207
T3 - Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
SP - 1151
EP - 1155
BT - Proceedings of the Eighth International Conference on Document Analysis and Recognition
T2 - 8th International Conference on Document Analysis and Recognition
Y2 - 31 August 2005 through 1 September 2005
ER -