Piotr Koniusz

Commonwealth Scientific and Industrial Research Organisation — Australia

Computer Science · Physical Sciences

35h-index4.4kcitations175works2.72yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Domain Adaptation and Few-Shot Learning(63), Human Pose and Action Recognition(43), Advanced Neural Network Applications(34), Advanced Image and Video Retrieval Techniques(33), Multimodal Machine Learning Applications(32)

Publications175 total

Exploiting Class-agnostic Visual Prior for Few-shot Keypoint Detection
International Journal of Computer Vision·2026Open Access
MMT-ARD: Multimodal Multi-Teacher Adversarial Distillation for Robust Vision-Language Models
arXiv (Cornell University)·2025Open Access
Graph Your Own Prompt
arXiv (Cornell University)·2025Open Access
Feature Hallucination for Self-supervised Action Recognition
International Journal of Computer Vision·2025· 1 citedOpen Access
Beyond decoders: Learning prompt-aware features for few-shot object counting
Neurocomputing·2025
Feature Hallucination for Self-supervised Action Recognition
arXiv (Cornell University)·2025Open Access
Noise Consistency Regularization for Improved Subject-Driven Image Synthesis
2025· 1 cited
BiLoRA: Almost-orthogonal Parameter Spaces for Continual Learning
2025· 1 cited
Open-World Objectness Modeling Unifies Novel Object Detection
2025· 2 cited
Noise Consistency Regularization for Improved Subject-Driven Image Synthesis
arXiv (Cornell University)·2025Open Access
WWW 2025 Workshop Chairs' Welcome
2025Open Access
Graph Self-Supervised Learning with Learnable Structural and Positional Encodings
2025Open Access
Stabilizing Modality Gap & Lowering Gradient Norms Improve Zero-Shot Adversarial Robustness of VLMs
2025Open Access
Understanding and Mitigating Hyperbolic Dimensional Collapse in Graph Contrastive Learning
2025· 4 citedOpen Access
Math Blind: Failures in Diagram Understanding Undermine Reasoning in MLLMs
arXiv (Cornell University)·2025Open Access
Inductive Graph Few-shot Class Incremental Learning
2025· 4 citedOpen Access
Point-PNG: Conditional Pseudo-Negatives Generation for Point Cloud Pre-Training
IEEE Access·2025Open Access
Adaptive Multi-head Contrastive Learning
Lecture notes in computer science·2024· 4 cited
OpenKD: Opening Prompt Diversity for Zero- and Few-Shot Keypoint Detection
Lecture notes in computer science·2024· 1 cited
When Spatial meets Temporal in Action Recognition
arXiv (Cornell University)·2024Open Access
page 1 of 9Next →

Frequent Co-authors

Hongguang Zhang(13), Hongdong Li(9), Krystian Mikolajczyk(6), Hao Zhu(6), Lei Wang(5), Mehrtash Harandi(4), Fatih Porikli(4), Yuchao Dai(3), Christian Simon(3), Richard Nock(3), Philip H. S. Torr(3), Anoop Cherian(3), Yifei Zhang(3), Irwin King(3), Lei Wang(3), Naila Murray(3), Hongguang Zhang(2), Li Zhang(2), Xiaojuan Qi(2), Yusuf Tas(2)