Results that are similar to your search

Similar results

Cover image for Scientific Image Forgery Detection — Kaggle Competition Part...
Scientific Image Forgery Detection — Kaggle Competition Participated in the ongoing Kaggle competition on Copy-Move Forgery Detection in Scientific Images, aimed at identifying manipulated biomedical figures that can compromise research integrity. For this challenge, I developed a ResNet50 + U-Net hybrid segmentation model using PyTorch, designed to detect and segment forged regions at the pixel level. My approach combines Dice and Focal losses for balanced training, WeightedRandomSampling to oversample forged images, and Test-Time Augmentation (TTA) to improve prediction robustness. Achieved an initial score of 0.303 on the public leaderboard. I’m continuing to experiment with architecture tuning, learning rate schedules, and other loss functions to further enhance performance and generalization.
0
31
Cover image for AI Legal Case Analysis — Expert Evidence Prediction This pro...
AI Legal Case Analysis — Expert Evidence Prediction This project demonstrates an AI tool that analyzes legal cases and predicts how a judge may view expert testimony. Using structured case data and LLM reasoning, the system compares past rulings involving Dr. Bailey with a new case scenario. By parsing the actual judgment text: “I accept Dr Bailey’s evidence…” The model evaluates patterns in credibility, cross-examination outcomes, and judicial language. It then generates a predicted ruling, concluding that the judge would likely consider Dr. Bailey’s expert evidence credible again. This showcases: Smart case-to-case comparison Expert-evidence credibility scoring LLM legal reasoning Structured JSON analysis #LegalTech #AIforLaw #CaseAnalysis #LLMReasoning #ContraPortfolio
23
183
Cover image for My first post here on Contra! A project from last year that ...
My first post here on Contra! A project from last year that I really enjoyed working on: Style Transfer using Transformers. The identity was inspired by mountains, factories, and the shared effort behind every great piece of art. The goal was to blend structural integrity with artistic styles, transforming everyday scenes into visually striking pieces while preserving their core essence. I leveraged Transformer-based architectures to achieve arbitrary style transfer, experimenting with different styles and datasets to create diverse outputs. This project challenged me to combine technical rigor with creative exploration, and it’s one of my favorite projects because of how the results turned out. Excited to start sharing more of my work here 🙌 and looking forward to connecting with other creators and enthusiasts!
0
16