Filters








752 Hits in 1.5 sec

Discriminative Bayesian Dictionary Learning for Classification [article]

Naveed Akhtar, Faisal Shafait, Ajmal Mian
2015 arXiv   pre-print
We propose a Bayesian approach to learn discriminative dictionaries for sparse representation of data. The proposed approach infers probability distributions over the atoms of a discriminative dictionary using a Beta Process. It also computes sets of Bernoulli distributions that associate class labels to the learned dictionary atoms. This association signifies the selection probabilities of the dictionary atoms in the expansion of class-specific data. Furthermore, the non-parametric character
more » ... the proposed approach allows it to infer the correct size of the dictionary. We exploit the aforementioned Bernoulli distributions in separately learning a linear classifier. The classifier uses the same hierarchical Bayesian model as the dictionary, which we present along the analytical inference solution for Gibbs sampling. For classification, a test instance is first sparsely encoded over the learned dictionary and the codes are fed to the classifier. We performed experiments for face and action recognition; and object and scene-category classification using five public datasets and compared the results with state-of-the-art discriminative sparse representation approaches. Experiments show that the proposed Bayesian approach consistently outperforms the existing approaches.
arXiv:1503.07989v1 fatcat:2etf4qhyhfbv5pg45t7aailrbe

A Statistical Approach to Signal Denoising Based on Data-driven Multiscale Representation [article]

Khuram Naveed, Muhammad Tahir Akhtar, Muhammad Faisal Siddiqui, Naveed ur Rehman
2020 arXiv   pre-print
We develop a data-driven approach for signal denoising that utilizes variational mode decomposition (VMD) algorithm and Cramer Von Misses (CVM) statistic. In comparison with the classical empirical mode decomposition (EMD), VMD enjoys superior mathematical and theoretical framework that makes it robust to noise and mode mixing. These desirable properties of VMD materialize in segregation of a major part of noise into a few final modes while majority of the signal content is distributed among
more » ... earlier ones. To exploit this representation for denoising purpose, we propose to estimate the distribution of noise from the predominantly noisy modes and then use it to detect and reject noise from the remaining modes. The proposed approach first selects the predominantly noisy modes using the CVM measure of statistical distance. Next, CVM statistic is used locally on the remaining modes to test how closely the modes fit the estimated noise distribution; the modes that yield closer fit to the noise distribution are rejected (set to zero). Extensive experiments demonstrate the superiority of the proposed method as compared to the state of the art in signal denoising and underscore its utility in practical applications where noise distribution is not known a priori.
arXiv:2006.00640v1 fatcat:kisbitkmv5dwhpv5oev6pmzbte

Geometric Feature Learning for 3D Meshes [article]

Huan Lei, Naveed Akhtar, Mubarak Shah, Ajmal Mian
2021 arXiv   pre-print
Naveed Akhtar is the recipient of Office of National Intelligence Postdoctoral Grant (project number NIPG-2021-001) funded by the Australian Government.  ... 
arXiv:2112.01801v1 fatcat:qz5hc6xvrba7picldsx7rprzru

Adversarial Attack on Skeleton-based Human Action Recognition [article]

Jian Liu, Naveed Akhtar, Ajmal Mian
2019 arXiv   pre-print
Deep learning models achieve impressive performance for skeleton-based human action recognition. However, the robustness of these models to adversarial attacks remains largely unexplored due to their complex spatio-temporal nature that must represent sparse and discrete skeleton joints. This work presents the first adversarial attack on skeleton-based action recognition with graph convolutional networks. The proposed targeted attack, termed Constrained Iterative Attack for Skeleton Actions
more » ... A), perturbs joint locations in an action sequence such that the resulting adversarial sequence preserves the temporal coherence, spatial integrity, and the anthropomorphic plausibility of the skeletons. CIASA achieves this feat by satisfying multiple physical constraints, and employing spatial skeleton realignments for the perturbed skeletons along with regularization of the adversarial skeletons with Generative networks. We also explore the possibility of semantically imperceptible localized attacks with CIASA, and succeed in fooling the state-of-the-art skeleton action recognition models with high confidence. CIASA perturbations show high transferability for black-box attacks. We also show that the perturbed skeleton sequences are able to induce adversarial behavior in the RGB videos created with computer graphics. A comprehensive evaluation with NTU and Kinetics datasets ascertains the effectiveness of CIASA for graph-based skeleton action recognition and reveals the imminent threat to the spatio-temporal deep learning tasks in general.
arXiv:1909.06500v1 fatcat:yo5nw2jsyfg3dgmqo4qzudjafu

Empirical Autopsy of Deep Video Captioning Frameworks [article]

Nayyer Aafaq, Naveed Akhtar, Wei Liu, Ajmal Mian
2019 arXiv   pre-print
Contemporary deep learning based video captioning follows encoder-decoder framework. In encoder, visual features are extracted with 2D/3D Convolutional Neural Networks (CNNs) and a transformed version of those features is passed to the decoder. The decoder uses word embeddings and a language model to map visual features to natural language captions. Due to its composite nature, the encoder-decoder pipeline provides the freedom of multiple choices for each of its components, e.g the choices of
more » ... Ns models, feature transformations, word embeddings, and language models etc. Component selection can have drastic effects on the overall video captioning performance. However, current literature is void of any systematic investigation in this regard. This article fills this gap by providing the first thorough empirical analysis of the role that each major component plays in a contemporary video captioning pipeline. We perform extensive experiments by varying the constituent components of the video captioning framework, and quantify the performance gains that are possible by mere component selection. We use the popular MSVD dataset as the test-bed, and demonstrate that substantial performance gains are possible by careful selection of the constituent components without major changes to the pipeline itself. These results are expected to provide guiding principles for future research in the fast growing direction of video captioning.
arXiv:1911.09345v1 fatcat:3jlncx5i45ctbob6p7ychpxivm

Intercropping Maize with Cowpeas and Mungbean under Rainfed Conditions

Naveed Akhtar ., Mirza Hassan ., Akhtar Ali ., Muhammad Riaz .
2000 Pakistan Journal of Biological Sciences  
Akhtar et al.: Intercropping, maize, cowpeas, mungbean, biological and economical efficient Singh (1981) .  ... 
doi:10.3923/pjbs.2000.647.648 fatcat:2bvr5fjlofg4nakirmwc5ctjau

Sparseness helps: Sparsity Augmented Collaborative Representation for Classification [article]

Naveed Akhtar and Faisal Shafait and Ajmal Mian
2015 arXiv   pre-print
Many classification approaches first represent a test sample using the training samples of all the classes. This collaborative representation is then used to label the test sample. It was a common belief that sparseness of the representation is the key to success for this classification scheme. However, more recently, it has been claimed that it is the collaboration and not the sparseness that makes the scheme effective. This claim is attractive as it allows to relinquish the computationally
more » ... ensive sparsity constraint over the representation. In this paper, we first extend the analysis supporting this claim and then show that sparseness explicitly contributes to improved classification, hence it should not be completely ignored for computational gains. Inspired by this result, we augment a dense collaborative representation with a sparse representation and propose an efficient classification method that capitalizes on the resulting representation. The augmented representation and the classification method work together meticulously to achieve higher accuracy and lower computational time compared to state-of-the-art collaborative representation based classification approaches. Experiments on benchmark face, object and action databases show the efficacy of our approach.
arXiv:1511.08956v1 fatcat:ecufmwd6enbtvcdcwjaawjmrwq

Sebumetric and mexametric evaluation of a fennel based cream

Akhtar Rasul, Naveed Akhtar, Muhammad Iqbal, Barkat Ali Khan, Asadullah Madni, Ghulam Murtaza, Muhammad Khurram Waqas, Tariq Mahmood
2012 ScienceAsia  
This study aims to investigate the effects of a topical cream (emulsion) containing 4% extract of fennel (Foeniculum vulgare) on functional skin parameters like sebum content, skin melanin, and skin erythaema, using the base without fennel as a control. Fennel extract was entrapped in the inner aqueous phase of emulsion. The creams were applied to 11 healthy male volunteers for a period of 12 weeks. Skin parameters were measured fortnightly using a Mexameter MPA 5 and a Sebumeter MPA 5 to
more » ... ine the effects produced. The base showed insignificant (p > 0.05) increase while the active formulation showed significant (p 0.05) decrease in skin melanin and sebum content, which can be attributed to the presence of linoleic acid and oleic acids in fennel. Analysis using a paired sample t-test showed that the base decreased skin erythaema insignificantly while the formulation exerted a significant decrease indicating that it possessed anti-erythaemic effects. The formulation can therefore be used safely for the treatment of acne and as a skin whitening agent in males.
doi:10.2306/scienceasia1513-1874.2012.38.262 fatcat:cvo3zeg75beqnghfmbc7h3m2qe

Attack to Fool and Explain Deep Networks [article]

Naveed Akhtar, Muhammad A. A. K. Jalwana, Mohammed Bennamoun, Ajmal Mian
2021 arXiv   pre-print
Naveed Akhtar. He is a recipient of scholarship for international research fees (SIRF).  ... 
arXiv:2106.10606v1 fatcat:y4s7lmaw7rcqhpzrwhfb5yvjmy

Boosting Deep Transfer Learning for COVID-19 Classification [article]

Fouzia Altaf, Syed M.S. Islam, Naeem K. Janjua, Naveed Akhtar
2021 arXiv   pre-print
COVID-19 classification using chest Computed Tomography (CT) has been found pragmatically useful by several studies. Due to the lack of annotated samples, these studies recommend transfer learning and explore the choices of pre-trained models and data augmentation. However, it is still unknown if there are better strategies than vanilla transfer learning for more accurate COVID-19 classification with limited CT data. This paper provides an affirmative answer, devising a novel 'model'
more » ... n technique that allows a considerable performance boost to transfer learning for the task. Our method systematically reduces the distributional shift between the source and target domains and considers augmenting deep learning with complementary representation learning techniques. We establish the efficacy of our method with publicly available datasets and models, along with identifying contrasting observations in the previous studies.
arXiv:2102.08085v1 fatcat:ceowqiriwnajrhwdmvenbnjykm

Controlled Caption Generation for Images Through Adversarial Attacks [article]

Nayyer Aafaq, Naveed Akhtar, Wei Liu, Mubarak Shah, Ajmal Mian
2021 arXiv   pre-print
Deep learning is found to be vulnerable to adversarial examples. However, its adversarial susceptibility in image caption generation is under-explored. We study adversarial examples for vision and language models, which typically adopt an encoder-decoder framework consisting of two major components: a Convolutional Neural Network (i.e., CNN) for image feature extraction and a Recurrent Neural Network (RNN) for caption generation. In particular, we investigate attacks on the visual encoder's
more » ... en layer that is fed to the subsequent recurrent network. The existing methods either attack the classification layer of the visual encoder or they back-propagate the gradients from the language model. In contrast, we propose a GAN-based algorithm for crafting adversarial examples for neural image captioning that mimics the internal representation of the CNN such that the resulting deep features of the input image enable a controlled incorrect caption generation through the recurrent network. Our contribution provides new insights for understanding adversarial attacks on vision systems with language component. The proposed method employs two strategies for a comprehensive evaluation. The first examines if a neural image captioning system can be misled to output targeted image captions. The second analyzes the possibility of keywords into the predicted captions. Experiments show that our algorithm can craft effective adversarial images based on the CNN hidden layers to fool captioning framework. Moreover, we discover the proposed attack to be highly transferable. Our work leads to new robustness implications for neural image captioning.
arXiv:2107.03050v1 fatcat:chzibor54vdhpoq7vt4uo5ejkm

Octree guided CNN with Spherical Kernels for 3D Point Clouds [article]

Huan Lei, Naveed Akhtar, Ajmal Mian
2019 arXiv   pre-print
We propose an octree guided neural network architecture and spherical convolutional kernel for machine learning from arbitrary 3D point clouds. The network architecture capitalizes on the sparse nature of irregular point clouds, and hierarchically coarsens the data representation with space partitioning. At the same time, the proposed spherical kernels systematically quantize point neighborhoods to identify local geometric structures in the data, while maintaining the properties of
more » ... nvariance and asymmetry. We specify spherical kernels with the help of network neurons that in turn are associated with spatial locations. We exploit this association to avert dynamic kernel generation during network training that enables efficient learning with high resolution point clouds. The effectiveness of the proposed technique is established on the benchmark tasks of 3D object classification and segmentation, achieving new state-of-the-art on ShapeNet and RueMonge2014 datasets.
arXiv:1903.00343v1 fatcat:lvkqi6izz5e4feslxnzpmxu3fu

COMPLICATED INGUINAL HERNIAS;

Naveed Akhtar, Syed Shams- Ul-Hassan, Muhammad Sabir, M. Nauman Ashraf
2019 The Professional Medical Journal  
Herniorrhaphy and hernioplasty are the two most common modalities used with different degree of success and complication rates in the treatment of inguinal hernia. Several studies show that use of mesh is superior to the non-mesh operations in inguinal hernia surgery.It is generally believed that the use of biomaterials should be limited to non-infected surgical fields.Now the concept regarding use of mesh in complicated hernias is changing as shown by many studies. Current study is being
more » ... d to observe the outcomes of the mesh hernioplasty in treatment of complicated inguinal hernias in emergency so that in future appropriate and safe technique may be suggested for repair of complicated hernias in emergency setting. Objectives: To compare the outcome of hernioplasty and herniorrhaphy in emergency for the treatment of complicated (Irreducible/obstructed) inguinal hernias regarding wound infection and hospital stay. Material & Methods:... Study Design: Randomized control trial. Setting: Surgical ward, Sheikh Zayed Hospital, Rahim yar khan. Period:09 months from 01-01-2016 to 30-09-2016. Sample Size: A total of 64 patients with 32 patients were included in each group, with confidence level of 95% and power of 80% and anticipated mean level of hospital stay in group 1 of 5±3.4 days versus 3±2.1 days in group 2. Sampling Technique: Non-probability, consecutive sampling. Results: In this study there were total 64 cases with 32 in each group. The mean age was 41.69±11.06 years and the mean duration of hernia obstruction was 12.83±4.97 hours. There was no significant difference in terms of age, duration of hernia and hernial obstruction between both groups. Seroma was seen in 5 (7.81%) out of 64 cases while wound infection was seen in 8 (12.50%) of cases. Seroma was seen in 2 (6.25%) out of 30 cases in herniorrhaphy as compared to 3 (9.38%) out of 32 cases with hernioplasty with p value of 0.64. Wound infection was seen in equally 4 (12.50%) out of 32 cases in both groups with p value of 1.0. Duration of hospital stay was 4.66±1.36 in patients with herniorrhaphy as compared to 4.53±1.37 days with hernioplasty with p value= 0.82. There was no significant difference in terms of age groups, duration of hernia and its obstruction between both groups regarding seroma. There was also no significant association among any of the confounding factors regarding the wound infection and length of the hospital stay between the both groups. Conclusion: We can perform hernioplasty as compared to herniorrhaphy for complicated inguinal hernia with similar complications and better success rates in the same emergency setting.
doi:10.29309/tpmj/2019.26.03.3232 fatcat:4m6mdvs3zba77lzs75fylg3zwe

A Survey of Neural Trojan Attacks and Defenses in Deep Learning [article]

Jie Wang, Ghulam Mubashar Hassan, Naveed Akhtar
2022 arXiv   pre-print
Artificial Intelligence (AI) relies heavily on deep learning - a technology that is becoming increasingly popular in real-life applications of AI, even in the safety-critical and high-risk domains. However, it is recently discovered that deep learning can be manipulated by embedding Trojans inside it. Unfortunately, pragmatic solutions to circumvent the computational requirements of deep learning, e.g. outsourcing model training or data annotation to third parties, further add to model
more » ... ility to the Trojan attacks. Due to the key importance of the topic in deep learning, recent literature has seen many contributions in this direction. We conduct a comprehensive review of the techniques that devise Trojan attacks for deep learning and explore their defenses. Our informative survey systematically organizes the recent literature and discusses the key concepts of the methods while assuming minimal knowledge of the domain on the readers part. It provides a comprehensible gateway to the broader community to understand the recent developments in Neural Trojans.
arXiv:2202.07183v1 fatcat:cmvnrimoofbgveg2btpu42ibeu

Orthogonal Deep Models As Defense Against Black-Box Attacks [article]

Mohammad A. A. K. Jalwana, Naveed Akhtar, Mohammed Bennamoun, Ajmal Mian
2020 arXiv   pre-print
TABLE Naveed Akhtar received his PhD in Computer Vision from The University of Western Australia (UWA) and Master degree in Computer Science from Hochschule Bonn-Rhein-Sieg, Germany.  ...  Later, Akhtar et. al [27] devised a label-universal technique to fool a model on an entire category of object in a targeted manner.  ... 
arXiv:2006.14856v1 fatcat:bkapm2p2mjd3xjfy3mxpvobsea
« Previous Showing results 1 — 15 out of 752 results