You'll be able to see their introduction and other information once they have accepted your connection request.
• Built a Keras implementation of Conditional Analogy GAN (CAGAN) for cloth-swapping. • Improved the original CAGAN via incorporating modules from image completion and image synthesis networks to generate more visually appealing results.
• Finished in 3rd place out of 217 teams worldwide for the 2.5-month competition held by SIGNATE and Fast Retailing. • Developed an effective deep learning approach combining hand-crafted and deep representations to classify the color on clothing featured in fashion images.
• Finished in top 11% out of 2293 teams at the worldwide Kaggle competition. • Implemented convolutional neural network in Keras to classifiy fish species.