Skip to content

Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs. / Ponce Lopez, Victor; Burghardt, Tilo; Sun, Yue; Hannuna, Sion; Aldamen, Dima; Mirmehdi, Majid.

International Conference on Image Analysis and Processing: Lecture Notes in Computer Science. Vol. 11751 Springer, 2019. p. 488-498 (Lecture Notes in Computer Science; Vol. 11751).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Ponce Lopez, V, Burghardt, T, Sun, Y, Hannuna, S, Aldamen, D & Mirmehdi, M 2019, Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs. in International Conference on Image Analysis and Processing: Lecture Notes in Computer Science. vol. 11751, Lecture Notes in Computer Science, vol. 11751, Springer, pp. 488-498. https://doi.org/10.1007/978-3-030-30642-7_44

APA

Ponce Lopez, V., Burghardt, T., Sun, Y., Hannuna, S., Aldamen, D., & Mirmehdi, M. (2019). Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs. In International Conference on Image Analysis and Processing: Lecture Notes in Computer Science (Vol. 11751, pp. 488-498). (Lecture Notes in Computer Science; Vol. 11751). Springer. https://doi.org/10.1007/978-3-030-30642-7_44

Vancouver

Ponce Lopez V, Burghardt T, Sun Y, Hannuna S, Aldamen D, Mirmehdi M. Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs. In International Conference on Image Analysis and Processing: Lecture Notes in Computer Science. Vol. 11751. Springer. 2019. p. 488-498. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-30642-7_44

Author

Ponce Lopez, Victor ; Burghardt, Tilo ; Sun, Yue ; Hannuna, Sion ; Aldamen, Dima ; Mirmehdi, Majid. / Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs. International Conference on Image Analysis and Processing: Lecture Notes in Computer Science. Vol. 11751 Springer, 2019. pp. 488-498 (Lecture Notes in Computer Science).

Bibtex

@inproceedings{d2c1eeff1a304cc690b39548a681e3bd,
title = "Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs",
abstract = "We present a dual-stream CNN that learns both appearance and facial features in tandem from still images and, after feature fusion, infers person identities. We then describe an alternative architecture of a single, lightweight ID-CondenseNet where a face detector-guided DC-GAN is used to generate distractor person images for enhanced training. For evaluation, we test both architectures on FLIMA, a new extension of an existing person re-identification dataset with added frame-by-frame annotations of face presence. Although the dual-stream CNN can outperform the CondenseNet approach on FLIMA, we show that the latter surpasses all state-of-the-art architectures in top-1 ranking performance when applied to the largest existing person re-identification dataset, MSMT17. We conclude that whilst re-identification performance is highly sensitive to the structure of datasets, distractor augmentation and network compression have a role to play for enhancing performance characteristics for larger scale applications.",
author = "{Ponce Lopez}, Victor and Tilo Burghardt and Yue Sun and Sion Hannuna and Dima Aldamen and Majid Mirmehdi",
year = "2019",
month = "9",
day = "2",
doi = "10.1007/978-3-030-30642-7_44",
language = "English",
isbn = "9783030306410",
volume = "11751",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "488--498",
booktitle = "International Conference on Image Analysis and Processing",

}

RIS - suitable for import to EndNote

TY - GEN

T1 - Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs

AU - Ponce Lopez, Victor

AU - Burghardt, Tilo

AU - Sun, Yue

AU - Hannuna, Sion

AU - Aldamen, Dima

AU - Mirmehdi, Majid

PY - 2019/9/2

Y1 - 2019/9/2

N2 - We present a dual-stream CNN that learns both appearance and facial features in tandem from still images and, after feature fusion, infers person identities. We then describe an alternative architecture of a single, lightweight ID-CondenseNet where a face detector-guided DC-GAN is used to generate distractor person images for enhanced training. For evaluation, we test both architectures on FLIMA, a new extension of an existing person re-identification dataset with added frame-by-frame annotations of face presence. Although the dual-stream CNN can outperform the CondenseNet approach on FLIMA, we show that the latter surpasses all state-of-the-art architectures in top-1 ranking performance when applied to the largest existing person re-identification dataset, MSMT17. We conclude that whilst re-identification performance is highly sensitive to the structure of datasets, distractor augmentation and network compression have a role to play for enhancing performance characteristics for larger scale applications.

AB - We present a dual-stream CNN that learns both appearance and facial features in tandem from still images and, after feature fusion, infers person identities. We then describe an alternative architecture of a single, lightweight ID-CondenseNet where a face detector-guided DC-GAN is used to generate distractor person images for enhanced training. For evaluation, we test both architectures on FLIMA, a new extension of an existing person re-identification dataset with added frame-by-frame annotations of face presence. Although the dual-stream CNN can outperform the CondenseNet approach on FLIMA, we show that the latter surpasses all state-of-the-art architectures in top-1 ranking performance when applied to the largest existing person re-identification dataset, MSMT17. We conclude that whilst re-identification performance is highly sensitive to the structure of datasets, distractor augmentation and network compression have a role to play for enhancing performance characteristics for larger scale applications.

U2 - 10.1007/978-3-030-30642-7_44

DO - 10.1007/978-3-030-30642-7_44

M3 - Conference contribution

SN - 9783030306410

VL - 11751

T3 - Lecture Notes in Computer Science

SP - 488

EP - 498

BT - International Conference on Image Analysis and Processing

PB - Springer

ER -