Defense against adversarial attacks on deep learning in computer vision

Research output

Research output per year 2018 2021 2024

• 16 Conference paper
• 14 Article
• 1 Review article

Research output per year

Research output per year

Search results

• 2024

  Efficient Hyperparameter Optimization with Adaptive Fidelity Identification

  Jiang, J., Wen, Z., Mansoor, A. & Mian, A., 19 Jun 2024, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEEE, p. 26181 10 p.

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Applicability 100%
  • Optimization Method 100%
  • Surrogate Model 100%
  • Learning Curve 100%
  • Resource Usage 100%
  14 Downloads (Pure)

• Faster-BNI: Fast Parallel Exact Inference on Bayesian Networks

  Jiang, J., Wen, Z., Mansoor, A. & Mian, A., Aug 2024, In: IEEE Transactions on Parallel and Distributed Systems. 35, 8, p. 1444-1455 12 p.

  Research output: Contribution to journal › Article › peer-review

  • Bayesian Networks 100%
  • Widespread Adoption 12%
  • Source Codes 12%
  • Tree Algorithm 12%
  • Parallelism 12%

• Fast Inference for Probabilistic Graphical Models

  Jiang, J., Wen, Z., Mansoor, A. & Mian, A., 2024, Proceedings of the 2024 USENIX Annual Technical Conference, ATC 2024. USENIX Association, p. 95-110 16 p.

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  • Graphical Model 100%
  • Inference System 25%
  • Computational Complexity 25%
  • Importance Sampling 25%
  • Source Codes 12%
• 2023

  Deep Reconstruction of 3-D Human Poses From Video

  Liu, J., Akhtar, N. & Mian, A., Jun 2023, In: IEEE Transactions on Artificial Intelligence. 4, 3, p. 497-510 14 p.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  File

  • Annotation 100%
  • Computational Modeling 66%
  • Recurrent Neural Network 33%
  • Constrained Environment 33%
  • Long Short-Term Memory Network 33%
  5 Citations (Scopus)

  407 Downloads (Pure)

• Language Model Agnostic Gray-Box Adversarial Attack on Image Captioning

  Aafaq, N., Akhtar, N., Liu, W., Shah, M. & Mian, A., 2023, In: IEEE Transactions on Information Forensics and Security. 18, p. 626-638 13 p.

  Research output: Contribution to journal › Article › peer-review

  • Language Modeling 100%
  • Adversarial Machine Learning 100%
  • Image Captioning 100%
  • Recurrent Neural Network 33%
  • Internal Representation 33%
  10 Citations (Scopus)

• SDFA: Structure Aware Discriminative Feature Aggregation for Efficient Human Fall Detection in Video

  Zahan, S., Hassan, G. M. & Mian, A., 1 Aug 2023, In: IEEE Transactions on Industrial Informatics. 19, 8, p. 8713-8721 9 p.

  Research output: Contribution to journal › Article › peer-review

  • Discriminative Feature 100%
  • Generalization Ability 100%
  • Privacy Concern 100%
  • Hardware Cost 100%
  • Motion Feature 100%
  5 Citations (Scopus)
• 2022

  Adversarial Attack on Skeleton-Based Human Action Recognition

  Liu, J., Akhtar, N. & Mian, A., 1 Apr 2022, In: IEEE Transactions on Neural Networks and Learning Systems. 33, 4, p. 1609-1622 14 p.

  Research output: Contribution to journal › Article › peer-review

  • Action Recognition 100%
  • Adversarial Machine Learning 100%
  • Graph Convolutional Network 66%
  • Action Sequence 33%
  • Deep Learning Model 33%
  32 Citations (Scopus)

• Attack to Fool and Explain Deep Networks

  Akhtar, N., Jalwana, M., Bennamoun, M. & Mian, A. S., 1 Oct 2022, In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 44, 10, p. 5980-5995 16 p.

  Research output: Contribution to journal › Article › peer-review

  • Visual Representation 100%
  • Image Synthesis 50%
  • Image Manipulation 50%
  • Human Perception 50%
  • Decision Boundary 50%
  14 Citations (Scopus)

• Target-Aware Holistic Influence Maximization in Spatial Social Networks

  Cai, T., Li, J., Mian, A., Li, R. H., Sellis, T. & Yu, J. X., 1 Apr 2022, In: IEEE Transactions on Knowledge and Data Engineering. 34, 4, p. 1993-2007 15 p.

  Research output: Contribution to journal › Article › peer-review

  • Diffusion Model 100%
  • Social Network 100%
  • User Interaction 33%
  • Textual Information 33%
  • Computational Cost 33%
  140 Citations (Web of Science)

• Transferable 3D Adversarial Textures using End-to-end Optimization

  Pestana Cardeno, C., Akhtar, N., Rahnavard, N., Shah, M. & Mian, A., 2022, IEEE/CVF Winter Conference on Applications of Computer Vision. IEEE, Institute of Electrical and Electronics Engineers, p. 727-736 10 p.

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • Adversarial Machine Learning 100%
  • Research Direction 50%
  • Lighting Condition 50%
• 2021

  Adversarial Attacks and Defense on Deep Learning Classification Models using YC_bC_r Color Images

  Pestana, C., Akhtar, N., Liu, W., Glance, D. & Mian, A., 18 Jul 2021, IJCNN 2021 - International Joint Conference on Neural Networks, Proceedings. IEEE, Institute of Electrical and Electronics Engineers, (Proceedings of the International Joint Conference on Neural Networks; vol. 2021-July).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Classification Models 100%
  • Deep Learning Method 100%
  • Adversarial Machine Learning 100%
  • Model Prediction 100%
  • Colour Space 100%
  4 Citations (Scopus)

  148 Downloads (Pure)

• Annotation Tool and Urban Dataset for 3D Point Cloud Semantic Segmentation

  Ibrahim, M., Akhtar, N., Wise, M. & Mian, A., 2021, In: IEEE Access. 9, p. 35984-35996 13 p., 9363898.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  • Annotation 100%
  • Point Cloud 100%
  • Image Segmentation 100%
  • Deep Learning Method 33%
  • Training Data 16%
  13 Citations (Scopus)

• Assistive signals for deep neural network classifiers

  Pestana, C., Liu, W., Glance, D., Owens, R. & Mian, A., Jun 2021, Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. USA: IEEE, Institute of Electrical and Electronics Engineers, p. 1221-1225 5 p. (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Deep Neural Network 100%
  • Neural Network 100%
  • Confidence Score 50%
  • Adversarial Machine Learning 50%
  • Conventional Method 50%
  5 Citations (Scopus)

  79 Downloads (Pure)

• Attacking Image Classifiers to Generate 3D Textures

  Pestana, C. & Mian, A., 2021, Proceedings - 2021 International Conference on 3D Vision, 3DV 2021. IEEE, Institute of Electrical and Electronics Engineers, p. 1115-1124 10 p. (Proceedings - 2021 International Conference on 3D Vision, 3DV 2021).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • Adversarial Machine Learning 100%
  • Deep Learning Method 33%
  • Image Synthesis 33%
  • Texture Mapping 33%
  • super resolution 33%

• CAMERAS: Enhanced Resolution And Sanity preserving Class Activation Mapping for image saliency

  Jalwana, M. A. A. K., Akhtar, N., Bennamoun, M. & Mian, A., 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, Institute of Electrical and Electronics Engineers, p. 16322-16331 10 p. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Class Activation Mapping 100%
  • Feature Map 100%
  • Research Direction 50%
  • Evaluation Metric 50%
  • Estimating Model 50%
  38 Citations (Scopus)

  158 Downloads (Pure)

• Deep Affinity Network for Multiple Object Tracking

  Sun, S., Akhtar, N., Song, H., Mian, A. S. & Shah, M., 1 Jan 2021, In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 43, 1, p. 104-119 16 p., 8766896.

  Research output: Contribution to journal › Article › peer-review

  • Tracking Object 100%
  • data association 75%
  • Object Detection 50%
  • Deep Learning Method 50%
  • Tracking Method 25%
  234 Citations (Scopus)

• Exploiting structured cnns for semantic segmentation of unstructured point clouds from lidar sensor

  Ibrahim, M., Akhtar, N., Ullah, K. & Mian, A., Sept 2021, In: Remote Sensing. 13, 18, 3621.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  • Point Cloud 100%
  • Image Segmentation 100%
  • Convolutional Neural Network 28%
  • Neural Network Model 14%
  • Neural Network Architecture 14%
  8 Citations (Scopus)

• High Definition LiDAR mapping of Perth CBD

  Ibrahim, M., Akhtar, N., Jalwana, M. A. A. K., Wise, M. & Mian, A., 2021, DICTA 2021 - 2021 International Conference on Digital Image Computing: Techniques and Applications. Zhou, J., Salvado, O., Sohel, F., Borges, P. V. K. & Wang, S. (eds.). IEEE, Institute of Electrical and Electronics Engineers, (DICTA 2021 - 2021 International Conference on Digital Image Computing: Techniques and Applications).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • Central Business District 100%
  • Gaussian Distribution 100%
  • Western Australia 33%
  • Normal Density Functions 33%
  • Error Propagation 33%
  9 Citations (Scopus)

• Multidimensional ground reaction forces and moments from wearable sensor accelerations via deep learning

  Johnson, W. R., Mian, A., Robinson, M. A., Verheul, J., Lloyd, D. G. & Alderson, J., Jan 2021, In: IEEE Transactions on Biomedical Engineering. 68, 1, p. 289-297 9 p., 9130158.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  • Growth Hormone Releasing Factor 100%
  • Wearable Sensor 100%
  • Reaction Force 100%
  • Deep Learning Method 100%
  • Ground Reaction Force 100%
  67 Citations (Scopus)

• Picasso: A CUDA-based Library for Deep Learning over 3D Meshes

  Lei, H., Akhtar, N. & Mian, A., 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, Institute of Electrical and Electronics Engineers, p. 13849-13859 11 p. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Deep Learning Method 100%
  • Representation Learning 50%
  • Resolution Reduction 50%
  • Gaussian Mixture 50%
  • Feature Extraction 50%
  9 Citations (Scopus)

  244 Downloads (Pure)

• Semantic attribute enriched storytelling from a sequence of images

  Malakan, Z. M., Hassan, G. M., Jalwana, M. A. A. K., Aafaq, N. & Mian, A., 23 Dec 2021, 2021 Digital Image Computing Techniques and Applications (DICTA). Zhou, J., Salvado, O., Sohel, F., Borges, P. V. K. & Wang, S. (eds.). USA: IEEE, Institute of Electrical and Electronics Engineers, 8 p.

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File
  3 Citations (Scopus)

  52 Downloads (Pure)

• Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds

  Lei, H., Akhtar, N. & Mian, A., 1 Oct 2021, In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 43, 10, p. 3664-3680 17 p., 9051667.

  Research output: Contribution to journal › Article › peer-review

  • Convolution 100%
  • Edge 100%
  • Neural Network 100%
  • Convolutional Neural Network 100%
  • Point Cloud 100%
  108 Citations (Web of Science)
• 2020

  Anchored vertex exploration for community engagement in social networks

  Cai, T., Li, J., Hasan Haldar, N. A., Mian, A., Yearwood, J. & Sellis, T., Apr 2020, Proceedings - 2020 IEEE 36th International Conference on Data Engineering, ICDE 2020. IEEE, Institute of Electrical and Electronics Engineers, p. 409-420 12 p. 9101364. (Proceedings - International Conference on Data Engineering; vol. 2020-April).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • User Engagement 100%
  • Social Network 100%
  • Efficient Algorithm 50%
  • Community Member 50%
  13 Citations (Scopus)

• Attack to Explain Deep Representation

  Jalwana, M. A. A. K., Akhtar, N., Bennamoun, M. & Mian, A., 2020, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. USA: IEEE, Institute of Electrical and Electronics Engineers, p. 9540-9549 10 p. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • Visual Feature 100%
  • Deep Learning Method 100%
  • Image Synthesis 100%
  • Image Manipulation 100%
  • Human Perception 100%
  8 Citations (Scopus)

• Orthogonal Deep Models as Defense against Black-Box Attacks

  Jalwana, M. A. A. K., Akhtar, N., Bennamoun, M. & Mian, A., 2020, In: IEEE Access. 8, p. 119744-119757 14 p., 9129688.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  • Deep Learning Method 100%
  • Adversarial Machine Learning 100%
  • Black-Box Attack 100%
  • Black Box 100%
  • Computer Vision Task 50%
  5 Citations (Scopus)

• Point attention network for semantic segmentation of 3D point clouds

  Feng, M., Zhang, L., Lin, X., Gilani, S. Z. & Mian, A., Nov 2020, In: Pattern Recognition. 107, 107446.

  Research output: Contribution to journal › Article › peer-review

  Open Access

  File

  • Point Cloud 100%
  • Image Segmentation 100%
  • Attention (Machine Learning) 100%
  • Convolution Layer 50%
  • Convolution Kernel 25%
  125 Citations (Scopus)

  123 Downloads (Pure)

• SEGGCN: Efficient 3D point cloud segmentation with fuzzy spherical kernel

  Lei, H., Akhtar, N. & Mian, A., 2020, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. USA: IEEE, Institute of Electrical and Electronics Engineers, p. 11608-11617 10 p. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  • Point Cloud 100%
  • Point Cloud Segmentation 100%
  • Convolution 100%
  • Fuzzy Clustering 50%
  • World Application 50%
  80 Citations (Scopus)
• 2019

  Skepxels: Spatio-temporal Image Representation of Human Skeleton Joints for Action Recognition

  Liu, J., Akhtar, N. & Mian, A., 2019, IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPR 2019): Augmented Human Humancentric Understanding and 2D/3D Synthesis. USA: Computer Vision Foundation

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

• Spatio-Temporal Dynamics and Semantic Attribute Enriched Visual Encoding for Video Captioning

  Aafaq, N., Akhtar, N., Liu, W., Gilani, S. Z. & Mian, A., Jun 2019, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR). USA: IEEE, Institute of Electrical and Electronics Engineers, p. 12479-12488 10 p. 8953475. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; vol. 2019-June).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  File

  • Visual Feature 100%
  • Visual Encoding 100%
  • Temporal Dynamic 100%
  • Temporal Semantics 100%
  • Convolutional Neural Network 75%
  189 Citations (Scopus)

  282 Downloads (Pure)

• Video description: A survey of methods, datasets, and evaluation metrics

  Aafaq, N., Mian, A., Liu, W., Gilani, S. Z. & Shah, M., 1 Oct 2019, In: ACM Computing Surveys. 52, 6, 115.

  Research output: Contribution to journal › Review article › peer-review

  Open Access

  • Evaluation Metric 100%
  • Language Modeling 100%
  • Deep Learning Method 50%
  • Early Childhood 50%
  • Statistics 50%
  128 Citations (Scopus)
• 2018

  Defense against Universal Adversarial Perturbations

  Akhtar, N., Liu, J. & Mian, A., 14 Dec 2018, IEEE Conference on Computer Vision and Pattern Recognition. United States: IEEE, Institute of Electrical and Electronics Engineers, p. 3389-3398 10 p. 8578455. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

  Research output: Chapter in Book/Conference paper › Conference paper › peer-review

  Open Access

  • Deep Learning Method 100%
  • Input/Output 50%
  • Targeted Network 50%
  • Synthetic Image 50%
  • Discrete Cosine Transform 50%
  147 Citations (Scopus)