Girish AjmeraFine Tuning YOLOv10 for Custom Object DetectionYOLO (You Only Look Once) is a series of object detection models known for real-time object detection with high performance and low…Jun 30Jun 30
Girish AjmeraSimpleNet: A Simple Network for Image Anomaly Detection and LocalizationIntroductionMay 11May 11
Girish AjmeraExponential Moving Average based Batch-Wise Normalization for Input data in Convolutional Neural…Normalization plays a crucial role in training Convolutional Neural Networks (CNNs) by ensuring that the input data is appropriately…Mar 23Mar 23
Girish AjmeraExploring Algorithms to Determine Points Inside or Outside a PolygonIn computational geometry, one of the fundamental problems is determining whether a point lies inside or outside a polygon. As such…Mar 91Mar 91
Girish AjmeraFeature Extraction of Images using GLCM (Gray Level Cooccurrence Matrix)Feature extraction plays a pivotal role in image processing and computer vision tasks. Images contain vast amounts of data, and extracting…Feb 15Feb 15
Girish AjmeraSynthetic Data Generation to handle class imbalance using MixUp approachClass imbalance is a common obstacle that plagues various image classification tasks. It arises when one class is significantly…Feb 4Feb 4
Girish AjmeraAutoEncoders + tSNE : Exploratory Data Analysis on unlabeled Image Dataset.In the world of deep learning, a significant challenge is to analyze image datasets without any labels. When working with unlabeled image…Jan 151Jan 151
Girish AjmeraBeyond Max Pooling: Adopting Strided Convolutions for Enhanced Anomaly DetectionMax pooling has been a staple of convolutional neural networks for decades. By reducing feature map size, it enables networks to learn…Jan 12Jan 12
Girish AjmeraCOCO: Bridging Gaps in Deep Learning Through Annotation StandardizationImagine a world where annotation data is trapped in a labyrinth of custom formats, each demanding unique parsing code. Sounds like a recipe…Jan 8Jan 8
Girish AjmeraUnmasking the Tiny Tumor: Perceptual Loss vs. Pixel Counting in Medical Image SegmentationPerceptual loss is a concept used in training deep learning models to compare the output image with the target image based on how humans…Jan 4Jan 4