arXiv subjects

Pranjal Sahu

Publications and source records attributed to Pranjal Sahu.

5 recordsLinked to original sources

arXiv2026

Example-based Robust Abnormality Detection with Minimal Annotations using Exemplar Med-DETR

Reducing annotation requirements remains a key challenge in developing robust medical object detectors. To address this, Vision-Language (VL) object detection methods leverage grounding text information to enable powerful zero-shot and few-shot object detectors in the natural image domain [1, 2, 3, 4]. However, transferring these methods to the medical domain is challenging due to the absence of comparable quality and quantity of the grounding data. Regardless, significant contextual and non-imaging information exists in medical images that remains underutilized. Few-shot learning (FSL) techniques partially address this limitation but struggle to general ize to unseen medical findings and require extensive retraining when new findings are introduced [5, 6]. To overcome these challenges, we extend our prior EM-DETR framework [7] and introduce a scalable FS detection approach designed for efficient abnormality detection in Chest X-Ray (CXR) images under minimal supervision. The proposed architecture incorporates exemplar-based feature generation and domain-aware contrastive optimization, enabling effective adaptation to novel disease findings without exhaustive retraining. Our method achieves near state-of-the-art (SOTA) detection performance using less than 10% of the annotated data, demonstrating its potential for practical, annotation-efficient clinical deployment across both proprietary and public CXR datasets.

cs.CV↗

arXiv2022

Offense Detection in Dravidian Languages using Code-Mixing Index based Focal Loss

Over the past decade, we have seen exponential growth in online content fueled by social media platforms. Data generation of this scale comes with the caveat of insurmountable offensive content in it. The complexity of identifying offensive content is exacerbated by the usage of multiple modalities (image, language, etc.), code-mixed language and more. Moreover, even after careful sampling and annotation of offensive content, there will always exist a significant class imbalance between offensive and non-offensive content. In this paper, we introduce a novel Code-Mixing Index (CMI) based focal loss which circumvents two challenges (1) code-mixing in languages (2) class imbalance problem for Dravidian language offense detection. We also replace the conventional dot product-based classifier with the cosine-based classifier which results in a boost in performance. Further, we use multilingual models that help transfer characteristics learnt across languages to work effectively with low resourced languages. It is also important to note that our model handles instances of mixed script (say usage of Latin and Dravidian-Tamil script) as well. To summarize, our model can handle offensive language detection in a low-resource, class imbalanced, multilingual and code-mixed setting.

cs.CL↗

arXiv2022

DeepMTL Pro: Deep Learning Based MultipleTransmitter Localization and Power Estimation

In this paper, we address the problem of Multiple Transmitter Localization (MTL). MTL is to determine the locations of potential multiple transmitters in a field, based on readings from a distributed set of sensors. In contrast to the widely studied single transmitter localization problem, the MTL problem has only been studied recently in a few works. MTL is of great significance in many applications wherein intruders may be present. E.g., in shared spectrum systems, detection of unauthorized transmitters and estimating their power are imperative to efficient utilization of the shared spectrum. In this paper, we present DeepMTL, a novel deep-learning approach to address the MTL problem. In particular, we frame MTL as a sequence of two steps, each of which is a computer vision problem: image-to-image translation and object detection. The first step of mage-to-image translation essentially maps an input image representing sensor readings to an image representing the distribution of transmitter locations, and the second object detection step derives precise locations of transmitters from the image of transmitter distributions. For the first step, we design our learning model Sen2Peak, while for the second step, we customize a state-of-the-art object detection model YOLO-cust. Using DeepMTL as a building block, we also develop techniques to estimate transmit power of the localized transmitters. We demonstrate the effectiveness of our approach via extensive large-scale simulations and show that our approach outperforms the previous approaches significantly (by 50% or more) in performance metrics including localization error, miss rate, and false alarm rate. Our method also incurs a very small latency. We evaluate our techniques over a small-scale area with real testbed data and the testbed results align with the simulation results.

cs.NI↗

arXiv2021

Transfer Learning for Oral Cancer Detection using Microscopic Images

Oral cancer has more than 83% survival rate if detected in its early stages, however, only 29% of cases are currently detected early. Deep learning techniques can detect patterns of oral cancer cells and can aid in its early detection. In this work, we present the first results of neural networks for oral cancer detection using microscopic images. We compare numerous state-of-the-art models via transfer learning approach and collect and release an augmented dataset of high-quality microscopic images of oral cancer. We present a comprehensive study of different models and report their performance on this type of data. Overall, we obtain a 10-15% absolute improvement with transfer learning methods compared to a simple Convolutional Neural Network baseline. Ablation studies show the added benefit of data augmentation techniques with finetuning for this task.

cs.CV↗

arXiv2019

Handcrafted vs Deep Learning Classification for Scalable Video QoE Modeling

Mobile video traffic is dominant in cellular and enterprise wireless networks. With the advent of diverse applications, network administrators face the challenge to provide high QoE in the face of diverse wireless conditions and application contents. Yet, state-of-the-art networks lack analytics for QoE, as this requires support from the application or user feedback. While there are existing techniques to map QoS to QoE by training machine learning models without requiring user feedback, these techniques are limited to only few applications, due to insufficient QoE ground-truth annotation for ML. To address these limitations, we focus on video telephony applications and model key artefacts of spatial and temporal video QoE. Our key contribution is designing content- and device-independent metrics and training across diverse WiFi conditions. We show that our metrics achieve a median 90% accuracy by comparing with mean-opinion-score from more than 200 users and 800 video samples over three popular video telephony applications -- Skype, FaceTime and Google Hangouts. We further extend our metrics by using deep neural networks, more specifically we use a combined CNN and LSTM model. We achieve a median accuracy of 95% by combining our QoE metrics with the deep learning model, which is a 38% improvement over the state-of-the-art well known techniques.

cs.MM↗