WSSSP-Net: Weakly supervised semantic segmentation plugin network for face anti-spoofing
Krzysztof J. Galus, Piotr Syga, Piotr Kawa
2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
2026
Profile
Piotr Kawa
PhD Student
Research groups
Piotr Kawa holds BSc and MSc in Computer Science and is currently PhD Student at Wroclaw University of Science and Technology. In his work he focuses on applications of machine learning in the selected aspects of information security. His research and professional interests focus on image and speech processing including topics like DeepFake detection and Text-To-Speech / Voice Conversion systems. His experience in commercial projects covers identity verification, information security (media watermarking), as well as signal processing (audio and video).
Education
2019
BSc in Computer Science
Wroclaw University of Science and Technology
2020
MSc in Computer Science (Computer Security)
Wroclaw University of Science and Technology
Publications(14)
From chatbot to validator
Aleksandra D Kawala-Sterniuk, Mariusz Pelc, Michał Podpora, Marek Baranowski, Piotr Rogala, Piotr Kawa, Paweł Piróg, Anna Romaniewska, Wojciech Rogala
A dual-agent strategy towards trustworthy on-premise conversational LLMs. [Setúbal] : SCITEPRESS - Science and Technology Publications, cop. 2026. s. 220-227.
2026
Zgłoszenie patentowe opublikowane
Piotr Kawa, Paweł J Narolski, Filip Zagórski
nr WO 2025/141302, opubl. 03.07.2025. Zgłosz. nr 141302 z 27.12.2023. A computer-implemented method for object liveness detection, a system for object liveness detection, a computer device and a computer-readable storage medium / Piotr Kawa, Paweł Narolski, Filip Zagórski. 36 s : 3 rys.
2025
As good as it KAN get: high-fidelity audio representation
Piotr Kawa, Piotr Syga, Patryk Marszałek, Maciej Rut, Przemysław Spurek
New York : Association for Computing Machinery, cop. 2025. s. 2063-2073.
2025
Are audio DeepFake detection models polyglots?
Bartłomiej M. Marek, Piotr Kawa, Piotr Syga
W: 5th Symposium on Security and Privacy in Speech Communication, Delft, The Netherlands, 16 August 2025 / eds. Ingo Siegert [i in.]. [B.m. : b.w.], 2025. s. 48-55.
2025
Replay attacks against audio deepfake detection
Piotr Kawa, Nicolas M. Müller, Wei Herng Choong, Adriana Stan, Aditya Tirumala Bukkapatnam, Karla Pizzi, Alexander Wagner, Philip Sperl
[B.m.] : International Speech Communication Association, cop. 2025. s. 2245-2249.
2025
MLAAD: The multi-language audio anti-spoofing dataset
Piotr Kawa, Piotr Syga, Nicolas M. Müller, Wei Herng Choong, Edresson Casanova, Eren Gölge, Thorsten Müller, Philip Sperl, Konstantin Böttinger
Danvers, MA : IEEE, cop. 2024. s. 1-7.
2024
A new approach to voice authenticity
Piotr Kawa, Nicolas M. Müller, Shen Hu, Matthias Neu, Jennifer Williams, Philip Sperl, Konstantin Böttinger
[B.m.] : International Speech Communication Association, cop. 2024. s. 2245-2249.
2024
Improved DeepFake detection using Whisper features
Piotr Kawa, Marcin Plata, Michał J Czuba, Piotr Szymański, Piotr Syga
INTERSPEECH 2023
2023
SpecRNet: Towards faster and more accessible audio DeepFake detection
Piotr Kawa, Marcin Plata, Piotr Syga
Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022
2022
Attack Agnostic Dataset: Towards Generalization and Stabilization of Audio DeepFake Detection
Piotr Kawa, Marcin Plata, Piotr Syga
Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
2022
Verify it yourself: a note on activation functions’ influence on fast DeepFake detection
SECRYPT 2021: PROCEEDINGS OF THE 18TH INTERNATIONAL CONFERENCE ON SECURITY AND CRYPTOGRAPHY
2021