arXiv subjects

Manoj Babu

Publications and source records attributed to Manoj Babu.

3 recordsLinked to original sources

arXiv2025

Cross-Model Transferability of Adversarial Patches in Real-time Segmentation for Autonomous Driving

Adversarial attacks pose a significant threat to deep learning models, particularly in safety-critical applications like healthcare and autonomous driving. Recently, patch based attacks have demonstrated effectiveness in real-time inference scenarios owing to their 'drag and drop' nature. Following this idea for Semantic Segmentation (SS), here we propose a novel Expectation Over Transformation (EOT) based adversarial patch attack that is more realistic for autonomous vehicles. To effectively train this attack we also propose a 'simplified' loss function that is easy to analyze and implement. Using this attack as our basis, we investigate whether adversarial patches once optimized on a specific SS model, can fool other models or architectures. We conduct a comprehensive cross-model transferability analysis of adversarial patches trained on SOTA Convolutional Neural Network (CNN) models such PIDNet-S, PIDNet-M and PIDNet-L, among others. Additionally, we also include the Segformer model to study transferability to Vision Transformers (ViTs). All of our analysis is conducted on the widely used Cityscapes dataset. Our study reveals key insights into how model architectures (CNN vs CNN or CNN vs. Transformer-based) influence attack susceptibility. In particular, we conclude that although the transferability (effectiveness) of attacks on unseen images of any dimension is really high, the attacks trained against one particular model are minimally effective on other models. And this was found to be true for both ViT and CNN based models. Additionally our results also indicate that for CNN-based models, the repercussions of patch attacks are local, unlike ViTs. Per-class analysis reveals that simple-classes like 'sky' suffer less misclassification than others. The code for the project is available at: https://github.com/p-shekhar/adversarial-patch-transferability

cs.CV↗

arXiv2021

Object shape error modelling and simulation during early design stage by morphing Gaussian Random Fields

Geometric and dimensional variations in objects are caused by inevitable uncertainties in manufacturing processes and often lead to product quality issues. Failing to model the effect object shape errors, i.e., geometric and dimensional errors of parts, early during design phase inhibits the ability to predict such quality issues; consequently leading to expensive design changes after freezing of design. State-of-Art methodologies for modelling and simulating object shape error have limited defect fidelity, data versatility, and designer centricity that prevent their effective application during early design phase. Overcoming these limitations a novel Morphing Gaussian Random Field (MGRF) methodology for object shape error modelling and simulation is presented in this paper. The MGRF methodology has (i) high defect fidelity and is capable of simulating various part defects including local and global deformations, and technological patterns; (ii) high data versatility and can effectively simulate non-ideal parts under the constraint of limited data availability and can utilise historical non-ideal part data; (iii) designer centric capabilities such as performing `what if?' analysis of practically relevant defects; and (iv) capability to generate non-ideal parts conforming to statistical form tolerance specification. The aforementioned capabilities enable MGRF methodology to accurately model and simulate the effect of object shape variations on product quality during the early design phase. This is achieved by first, modelling the spatial correlation in the deviations of the part from its design nominal using Gaussian Random Field and then, utilising the modelled spatial correlations to generate non-ideal parts by conditional simulations. Practical applications of developed MGRF methodology and its advantages are demonstrated using sport-utility-vehicle door parts.

cs.CE↗

arXiv2020

An Online Learning Methodology for Performance Modeling of Graphics Processors

Approximately 18 percent of the 3.2 million smartphone applications rely on integrated graphics processing units (GPUs) to achieve competitive performance. Graphics performance, typically measured in frames per second, is a strong function of the GPU frequency, which in turn has a significant impact on mobile processor power consumption. Consequently, dynamic power management algorithms have to assess the performance sensitivity to the frequency accurately to choose the operating frequency of the GPU effectively. Since the impact of GPU frequency on performance varies rapidly over time, there is a need for online performance models that can adapt to varying workloads. This paper presents a light-weight adaptive runtime performance model that predicts the frame processing time of graphics workloads at runtime without apriori characterization. We employ this model to estimate the frame time sensitivity to the GPU frequency, i.e., the partial derivative of the frame time with respect to the GPU frequency. The proposed model does not rely on any parameter learned offline. Our experiments on commercial platforms with common GPU benchmarks show that the mean absolute percentage error in frame time and frame time sensitivity prediction are 4.2 and 6.7 percent, respectively.

eess.SY↗