Publications

Preprints

S. Ramasinghe, V. Shevchenko, G. Avraham, Y. Zuo, M. Abdolshah, H. Husain, and T. Ajanthan, A Visual Codebook for Robust Arbitrary Scale Super-Resolution, 2024.

Y. Duan, S. Ramasinghe, A. Long, S. Gould, and T. Ajanthan, Contrastive Difference Alignment Between Vision and Language for Composed Image Retrieval, 2024.

Y. Duan, S. Ramasinghe, S. Gould, and T. Ajanthan, Scaling Prompt Instructed Zero Shot Composed Image Retrieval with Image-Only Data, 2024.

X. Zhu, T. Ajanthan, C. Xu, D. Tao, A. van den Hengel, and V. Nguyen, Sample and Label Aware Augmentation for Tabular Data, 2023.

Journals

[new] Y. Lu, S. Gould, and T. Ajanthan, Bidirectionally Self-Normalizing Neural Networks, NN, 2023. [pdf] [arxiv] [talk] [bib]

@article{lu_bsnn_nn23,
  author = {Lu, Yao and Gould, Stephen and Ajanthan, Thalaiyasingam},
  title = {Bidirectionally Self-Normalizing Neural Networks},
  journal = {NN},
  year = {2023}
}

#DLT, #SPNN

R. de Bem, A. Ghosh, T. Ajanthan, O. Miksik, A. Boukhayma, N. Siddharth, and P. H. S. Torr, DGPose: Deep Generative Models for Human Body Analysis, IJCV, 2020. [pdf] [arxiv] [bib]

@article{debem_dgpose_ijcv2020,
  title={{DGPose:} Deep Generative Models for Human Body Analysis.},
  author={De Bem, Rodrigo and Ghosh, Arnab and Ajanthan, Thalaiyasingam and Miksik, Ondrej and Boukhayma, Adnane and Narayanaswamy, Siddharth and Torr, Philip HS},
  journal={IJCV},
  year={2020}
}

#GM

T. Joy*, A. Desmaison*, T. Ajanthan*, R. Bunel, M. Salzmann, P. Kohli, P. H. S. Torr, and M. P. Kumar, Efficient Relaxations for Dense CRFs with Sparse Higher Order Potentials, SIIMS, 2019. [pdf] [arxiv] [code] [bib]

@article{joy_higherorder_siam2018,
  title={Efficient Relaxations for Dense {CRFs} with Sparse Higher Order Potentials},
  author={Joy, Thomas and Desmaison, Alban and Ajanthan, Thalaiyasingam and Bunel, Rudy and Salzmann, Mathieu and Kohli, Pushmeet and Torr, Philip HS and Kumar, M Pawan},
  journal={SIIMS},
  year={2019}
}

#MRF

T. Ajanthan, R. Hartley, and M. Salzmann, Memory Efficient Max Flow for Multi-label Submodular MRFs, PAMI, 2018. [pdf] [arxiv] [talk] [code] [bib]

@article{ajanthan_memf_pami2018,
  title={Memory Efficient Max Flow for Multi-label Submodular {MRFs}},
  author={Ajanthan, Thalaiyasingam and Hartley, Richard and Salzmann, Mathieu},
  journal={PAMI},
  year={2018}
}

#MRF

Conferences

[new] E. Markou, T. Ajanthan, and S. Gould, Guiding Neural Collapse: Optimising Towards the Nearest Simplex Equiangular Tight Frame, NeurIPS, 2024. [to appear] [bib]

@article{markou_gnc_neurips24,
  author = {Markou, Evan, and Ajanthan, Thalaiyasingam, and Gould, Stephen},
  title = {Guiding Neural Collapse: Optimising Towards the Nearest Simplex Equiangular Tight Frame},
  journal = {NeurIPS},
  year = {2024}
}

#DLT

[new] S. Ramasinghe, V. Shevchenko, G. Avraham, and T. Ajanthan, Accept the Modality Gap: An Exploration in the Hyperbolic Space, CVPR, 2024. (highlight) [pdf] [talk] [code] [bib]

@article{ramasinghe_atmg_cvpr24,
  author = {Ramasinghe, Sameera, and Shevchenko, Violetta, and Avraham, Gil, and Ajanthan, Thalaiyasingam},
  title = {Accept the Modality Gap: An Exploration in the Hyperbolic Space},
  journal = {CVPR},
  year = {2024}
}

#VL

[new] P. Li, P. Purkait, T. Ajanthan, M. Abdolshah, R. Garg, H. Husain, C. Xu, S. Gould, W. Ouyang, and A. van den Hengel, Semi-Supervised Semantic Segmentation under Label Noise via Diverse Learning Groups, ICCV, 2023. [pdf] [supp] [bib]

@article{li_dlg_iccv23,
  author = {Li, Peixia and Purkait, Pulak and Ajanthan, Thalaiyasingam and Abdolshah, Majid and Garg, Ravi and Husain, Hisham and Xu, Chenchen and Gould, Stephen and Ouyang, Wanli and van den Hengel, Anton},
  title = {Semi-Supervised Semantic Segmentation under Label Noise via Diverse Learning Groups},
  journal = {ICCV},
  year = {2023}
}

#CV

A. Long, W. Yin, T. Ajanthan, V. Nguyen, P. Purkait, R. Garg. A. Blair, C. Shen, and A. van den Hengel, Retrieval Augmented Classification for Long-tail Visual Recognition, CVPR, 2022. [pdf] [arxiv] [bib]

@article{long_rac_cvpr22,
  author = {Long, Alexander and Yin, Wei and Ajanthan, Thalaiyasingam and Nguyen, Vu and Purkait, Pulak and Garg, Ravi and Blair, Alan and Shen, Chunhua and van den Hengel, Anton},
  title = {Retrieval Augmented Classification for Long-Tail Visual Recognition},
  journal = {CVPR},
  year = {2022}
}

#CV

K. Gupta, and T. Ajanthan, Improved Gradient based Adversarial Attacks for Quantized Networks, AAAI, 2022. [pdf] [arxiv] [bib]

@article{gupta_iga_aaai22,
  author = {Gupta, Kartik and Ajanthan, Thalaiyasingam},
  title = {Improved Gradient based Adversarial Attacks for Quantized Networks},
  journal = {AAAI},
  year = {2022}
}

#NNQ, #SPNN

A. Shaban*, A. Rahimi*, T. Ajanthan, B. Boots, and R. Hartley, Few-shot Weakly-Supervised Object Detection via Directional Statistics, WACV, 2022. [pdf] [supp] [arxiv] [bib]

@article{rahimi_wsod_eccv20,
  author = {Shaban, Amirreza, and Rahimi, Amir, and Ajanthan, Thalaiyasingam, and Boots, Byron, and Hartley, Richard},
  title = {Few-shot Weakly-Supervised Object Detection via Directional Statistics},
  journal = {WACV},
  year = {2022}
}

#CV

M. Sasdelli, T. Ajanthan, T. J. Chin, and G. Carneiro, A Chaos Theory Approach to Understand Neural Network Optimization, DICTA, 2021. [pdf] [bib]

@article{sasdelli_chaos_dicta21,
  author = {Sasdelli, Michele, and Ajanthan, Thalaiyasingam, and Chin, Tat-Jun, and Carneiro, Gustavo},
  title = {A Chaos Theory Approach to Understand Neural Network Optimization},
  journal = {DICTA},
  year = {2021}
}

T. Ajanthan*, K. Gupta*, P. H. S. Torr, R. Hartley, and P. K. Dokania, Mirror Descent View for Neural Network Quantization, AISTATS, 2021. [pdf] [supp] [arxiv] [code] [bib]

@article{ajanthan_mdnnq_aistats21,
  author = {Ajanthan, Thalaiyasingam, and Gupta, Kartik, and Torr, Philip HS, and Hartley, Richard and Dokania, Puneet K},
  title = {Mirror Descent View for Neural Network Quantization},
  journal = {AISTATS},
  year = {2021}
}

#NNQ

K. Gupta, A, Rahimi, T. Ajanthan, T. Mensink, C. Sminchisescu, and R. Hartley, Calibration of Neural Networks using Splines, ICLR, 2021. [pdf] [arxiv] [bib]

@article{gupta_spline_iclr21,
  author = {Gupta, Kartik and Rahimi, Amir and Ajanthan, Thalaiyasingam and Mensink, Thomas and Sminchisescu and Cristian and Hartley, Richard},
  title = {Calibration of Neural Networks using Splines},
  journal = {ICLR},
  year = {2021}
}

#NNC

N. Lee, T. Ajanthan, P. H. S. Torr, and M. Jaggi, Understanding the Effects of Data Parallelism and Sparsity on Neural Network Training, ICLR, 2021. [pdf] [arxiv] [bib]

@article{lee_dataparallelism_iclr21,
  author = {Lee, Namhoon and Ajanthan, Thalaiyasingam and Torr, Philip HS and Jaggi, Martin},
  title = {Understanding the Effects of Data Parallelism and Sparsity on Neural Network Training},
  journal = {ICLR},
  year = {2021}
}

#NNP

Z. Xu, T. Ajanthan, V. Vineet, and R. Hartley, RANP: Resource Aware Neuron Pruning at Initialization for 3D CNNs, 3DV, 2020. (oral, best student paper) [pdf] [arxiv] [talk] [code] [bib]

@article{xu_ranp_3dv20,
  author = {Xu, Zhiwei, and Ajanthan, Thalaiyasingam, and Vineet, Vibhav, and Hartley, Richard},
  title = {RANP: Resource Aware Neuron Pruning at Initialization for 3D CNNs},
  journal = {3DV},
  year = {2020}
}

#NNP

Z. Xu, T. Ajanthan, and R. Hartley, Fast and Differentiable Message Passing on Pairwise Markov Random Fields, ACCV, 2020. (oral) [pdf] [arxiv] [talk] [code] [bib]

@article{xu_trwp_accv20,
  author = {Xu, Zhiwei, and Ajanthan, Thalaiyasingam, and Hartley, Richard},
  title = {Fast and Differentiable Message Passing on Pairwise Markov Random Fields},
  journal = {ACCV},
  year = {2020}
}

#MRF

A. Rahimi*, A. Shaban*, T. Ajanthan, R. Hartley, and B. Boots, Pairwise Similarity Knowledge Transfer for Weakly Supervised Object Localization, ECCV, 2020. [pdf] [arxiv] [bib]

@article{rahimi_wsod_eccv20,
  author = {Rahimi, Amir, and Shaban, Amirreza, and Ajanthan, Thalaiyasingam, and Hartley, Richard, Boots, Byron},
  title = {Pairwise Similarity Knowledge Transfer for Weakly Supervised Object Localization},
  journal = {ECCV},
  year = {2020}
}

#CV

N. Lee, T. Ajanthan, S. Gould, and P. H. S. Torr, A Signal Propagation Perspective for Pruning Neural Networks at Initialization, ICLR, 2020. (spotlight) [pdf] [arxiv] [talk] [code] [bib]

@article{lee_disnip_iclr20,
  author = {Lee, Namhoon and Ajanthan, Thalaiyasingam and Gould, Stephen and Torr, Philip HS},
  title = {A Signal Propagation Perspective for Pruning Neural Networks at Initialization},
  journal = {ICLR},
  year = {2020}
}

#NNP, #SPNN

T. Ajanthan, P. K. Dokania, R. Hartley, and P. H. S. Torr, Proximal Mean-field for Neural Network Quantization, ICCV, 2019. [pdf] [supp] [arxiv] [poster] [talk] [code] [bib]

@article{ajanthan_pmf_iccv19,
  author = {Ajanthan, Thalaiyasingam, and Dokania, Puneet K and Hartley, Richard and Torr, Philip HS},
  title = {Proximal Mean-field for Neural Network Quantization},
  journal = {ICCV},
  year = {2019}
}

#NNQ

Alessio Tonioni, Oscar Rahnama*, Thomas Joy*, Luigi Di Stefano, T. Ajanthan, and P. H. S. Torr, Learning to Adapt for Stereo, CVPR, 2019. [pdf] [supp] [arxiv] [code] [bib]

@article{tonioni_l2a_cvpr19,
  author = {Tonioni, Alessio and Rahnama, Oscar and Joy, Thomas and Stefano, Luigi Di and Ajanthan, Thalaiyasingam and Torr, Philip HS},
  title = {Learning to Adapt for Stereo},
  journal = {CVPR},
  year = {2019}
}

#META

N. Lee, T. Ajanthan, and P. H. S. Torr, SNIP: Single-shot Network Pruning based on Connection Sensitivity, ICLR, 2019. [pdf] [arxiv] [poster] [talk] [code] [bib]

@article{lee_snip_iclr19,
  author = {Lee, Namhoon and Ajanthan, Thalaiyasingam and Torr, Philip HS},
  title = {{SNIP:} Single-shot Network Pruning based on Connection Sensitivity},
  journal = {ICLR},
  year = {2019}
}

#NNP

R. de Bem, A. Ghosh, A. Boukhayma, T. Ajanthan, N. Siddharth, and P. H. S. Torr, A Conditional Deep Generative Model of People in Natural Images, WACV, 2019. [pdf] [bib]

@article{debem_dgpose_wacv2019,
  title={A Conditional Deep Generative Model of People in Natural Images},
  author={De Bem, Rodrigo and Ghosh, Arnab and Boukhayma, Adnane and Ajanthan, Thalaiyasingam and and Narayanaswamy, Siddharth and Torr, Philip HS},
  journal={WACV},
  year={2019}
}

#GM

A. Chaudhry*, P. K. Dokania*, T. Ajanthan*, and P. H. S. Torr, Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence, ECCV, 2018. [pdf] [supp] [arxiv] [poster] [talk] [bib]

@article{chaudhry_rwalk_eccv2018,
  title={Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence},
  author={Chaudhry, Arslan and Dokania, Puneet K and Ajanthan, Thalaiyasingam and Torr, Philip HS},
  journal={ECCV},
  year={2018}
}

#IL

T. Ajanthan, A. Desmaison, R. Bunel, M. Salzmann, P. H. S. Torr, and M. P. Kumar, Efficient Linear Programming for Dense CRFs, CVPR, 2017. [pdf] [supp] [arxiv] [poster] [talk] [code] [bib]

@article{ajanthan_proxlp_cvpr2017,
  title={Efficient linear programming for dense {CRFs}},
  author={Ajanthan, Thalaiyasingam and Desmaison, Alban and Bunel, Rudy and Salzmann, Mathieu and Torr, Philip HS and Kumar, M Pawan},
  journal={CVPR},
  year={2017}
}

#MRF

T. Ajanthan, R. Hartley, and M. Salzmann, Memory Efficient Max Flow for Multi-label Submodular MRFs, CVPR, 2016. [pdf] [supp] [poster] [talk] [code] [bib]

@article{ajanthan_memf_cvpr2016,
  title={Memory Efficient Max Flow for Multi-label Submodular {MRFs}},
  author={Ajanthan, Thalaiyasingam and Hartley, Richard and Salzmann, Mathieu},
  journal={CVPR},
  year={2016}
}

#MRF

T. Ajanthan, R. Hartley, M. Salzmann, and H. Li, Iteratively Reweighted Graph Cut for Multi-label MRFs with Non-convex Priors, CVPR, 2015. [pdf] [arxiv] [poster] [code] [bib]

@article{ajanthan_irgc_cvpr2015,
  title={Iteratively Reweighted Graph Cut for Multi-label {MRFs} with Non-convex Priors},
  author={Ajanthan, Thalaiyasingam and Hartley, Richard and Salzmann, Mathieu and Li, Hongdong},
  journal={CVPR},
  year={2015}
}

#MRF

T. Ajanthan, P Kamalaruban, and R. Rodrigo, Automatic Number Plate Recognition in Low Quality Videos, ICIIS, 2013. [pdf] [bib]

@article{ajanthan_anpr_iciis2013,
  title={Automatic Number Plate Recognition in Low Quality Videos},
  author={Ajanthan, Thalaiyasingam and Kamalaruban, Parameswaran and Rodrigo, Ranga},
  journal={ICIIS},
  year={2013}
}

Workshops

[new] Y. Liu, T. Ajanthan, H. Husain, and V. Nguyen, Self-Supervision Improves Diffusion Models for Tabular Data Imputation, ICLR Workshop: Generative Models for Decision Making, 2024. [pdf] [bib]

@article{liu_ssldm_iclr24,
  author = {Liu, Yixin, and Ajanthan, Thalaiyasingam, and Husain, Hisham, and Nguyen, Vu},
  title = {Self-Supervision Improves Diffusion Models for Tabular Data Imputation},
  journal = {ICLR Workshop: Generative Models for Decision Making},
  year = {2024}
}

[new] A. Long, T. Ajanthan, and A. van den Hengel, Modality-Aware Adaptation of Contrastive Language-Image Models, ICLR Workshop: Mathematical and Empirical Understanding of Foundation Models, 2023. [pdf] [bib]

@article{long_mater_iclr23,
  author = {Long, Alexander, and Ajanthan, Thalaiyasingam, and van den Hengel, Anton},
  title = {Modality-Aware Adaptation of Contrastive Language-Image Models},
  journal = {ICLR Workshop: Mathematical and Empirical Understanding of Foundation Models},
  year = {2023}
}

#VL

K. Gupta, T. Ajanthan, A. van den Hengel, and S. Gould, Understanding and Improving the Role of Projection Head in Self-Supervised Learning, NeurIPS Workshop: Self-Supervised Learning - Theory and Practice, 2022. [pdf] [bib]

@article{gupta_ssl_neurips22,
  author = {Gupta, Kartik, and Ajanthan, Thalaiyasingam, and van den Hengel, Anton, and Gould, Stephen},
  title = {Understanding and Improving the Role of Projection Head in Self-Supervised Learning},
  journal = {NeurIPS Workshop: Self-Supervised Learning - Theory and Practice},
  year = {2022}
}

D. D. Denipitiyage, T. Ajanthan, P. Kamalaruban, and A. Weller, Provable Defense Against Clustering Attacks on 3D Point Clouds, AAAI Workshop: Adversarial Machine Learning and Beyond, 2022. [pdf] [bib]

@article{denipitiyage_pcrobust_aaai22,
  author = {Denipitiyage, Dishanika Dewani and Ajanthan, Thalaiyasingam and Kamalaruban, Parameswaran and Weller, Adrian},
  title = {Provable Defense Against Clustering Attacks on 3D Point Clouds},
  journal = {AAAI Workshop on Adversarial Machine Learning and Beyond},
  year = {2022}
}

#CV

K. Gupta, and T. Ajanthan, Improved Gradient based Adversarial Attacks for Quantized Networks, CVPR Workshop: Adversarial Machine Learning in Computer Vision, 2020. (oral) [pdf] [arxiv] [bib]

@article{gupta_iga_cvpr20,
  author = {Gupta, Kartik and Ajanthan, Thalaiyasingam},
  title = {Improved Gradient based Adversarial Attacks for Quantized Networks},
  journal = {CVPR Workshop: Adversarial Machine Learning in Computer Vision},
  year = {2020}
}

#NNQ, #SPNN

A. Chaudhry, M. Rohrbach, M. Elhoseiny, T. Ajanthan, P. K. Dokania, P. H. S. Torr, and M. Ranzato, Continual Learning with Tiny Episodic Memories, ICML Workshop: Multi-Task and Lifelong Reinforcement Learning, 2019. [pdf] [arxiv] [bib]

@article{Chaudhry_tinyer_icml2019,
  title={Continual Learning with Tiny Episodic Memories},
  author={Chaudhry, Arslan and Rohrbach, Marcus and Elhoseiny, Mohamed and Ajanthan, Thalaiyasingam and Dokania, Puneet K and Torr, Philip HS and Ranzato, Marc'Aurelio},
  journal={ICML Workshop: Multi-Task and Lifelong Reinforcement Learning},
  year={2019}
}

#IL

R. de Bem, A. Ghosh, T. Ajanthan, O. Miksik, N. Siddharth, and P. H. S. Torr, A Semi-supervised Deep Generative Model for Human Body Analysis, ECCV Workshop: Human Behaviour Understanding, 2018. (oral) [pdf] [bib]

@article{debem_dgpose_eccv2018,
  title={A Semi-supervised Deep Generative Model for Human Body Analysis},
  author={De Bem, Rodrigo and Ghosh, Arnab and Ajanthan, Thalaiyasingam and Miksik, Ondrej and Narayanaswamy, Siddharth and Torr, Philip HS},
  journal={ECCV Workshop on Human Behaviour Understanding},
  year={2018}
}

#GM

M. Najafi*, V. Kulharia*, T. Ajanthan, and P. H. S. Torr, Similarity Learning for One-Shot Video Object Segmentation, CVPR Workshop: DAVIS Challenge on Video Object Segmentation, 2018. [pdf] [bib]

@article{najafi_vobjseg_davis2018,
  title={Similarity Learning for One-Shot Video Object Segmentation},
  author={Najafi, Mohammad and Kulharia, Viveka and Ajanthan, Thalaiyasingam and Torr, Philip HS},
  journal={CVPR Workshop: DAVIS Challenge on Video Object Segmentation},
  year={2018}
}

#META, #CV

Thesis

T. Ajanthan, Optimization of Markov Random Fields in Computer Vision, PhD Thesis, 2017. [pdf] [talk] [bib]

@article{ajanthan_thesis_2017,
  title={Optimization of {M}arkov Random Fields in Computer Vision},
  author={Ajanthan, Thalaiyasingam},
  journal={ANU PhD Thesis},
  year={2017}
}

#MRF

Technical Reports

T. Ajanthan, M. Ma, A. van den Hengel, and S. Gould, Adaptive Cross Batch Normalization for Metric Learning, arxiv:2303.17127, 2023. [pdf] [arxiv] [bib]

@article{ajanthan_axbn_arxiv23,
  author = {Ajanthan, Thalaiyasingam, and Ma, Matt, and van den Hengel, Anton, and Gould, Stephen},
  title = {Adaptive Cross Batch Normalization for Metric Learning},
  journal = {CoRR},
  year = {2023}
}

#CV

Z. Xu, T. Ajanthan, R. Hartley, Refining Semantic Segmentation with Superpixel by Transparent Initialization and Sparse Encoder, arxiv:2010.04363, 2020. [pdf] [arxiv] [bib]

@article{xu_supseg_arxiv20,
  author = {Xu, Zhiwei, and Ajanthan, Thalaiyasingam, and Hartley, Richard},
  title = {Refining Semantic Segmentation with Superpixel by Transparent Initialization and Sparse Encoder},
  journal = {CoRR},
  year = {2020}
}

#CV

A, Rahimi, K. Gupta, T. Ajanthan, T. Mensink, C. Sminchisescu, and R. Hartley, Post-hoc Calibration of Neural Networks, arxiv:2006.12807, 2020. [pdf] [arxiv] [bib]

@article{rahimi_glayers_arxiv20,
  author = {Rahimi, Amir and Gupta, Kartik and Ajanthan, Thalaiyasingam and Mensink, Thomas and Sminchisescu and Cristian and Hartley, Richard},
  title = {Post-hoc Calibration of Neural Networks},
  journal = {CoRR},
  year = {2020}
}

#NNC

R. Hartley, and T. Ajanthan, Generalized Range Moves, arXiv:1811.09171, 2018. [pdf] [arxiv] [bib]

@article{hartley_gswap_arxiv18,
  author = {Hartley, Richard and Ajanthan, Thalaiyasingam},
  title = {Generalized Range Moves},
  journal = {CoRR},
  year = {2018}
}

#MRF