NeuralTalk2

NeuralTalk2 is a Torch-based image-captioning system that generates natural-language descriptions for images with neural networks. It improves on the original NeuralTalk implementation through batching, GPU acceleration, and a more efficient training pipeline. The model combines convolutional neural network image features with a recurrent neural network language model. It supports fine-tuning the underlying CNN instead of relying only on fixed visual features. Training and evaluation tools are included for datasets such as MS COCO. The repository also contains visualization utilities, checkpoint conversion tools, and experimental video-captioning code. It is now primarily preserved as an educational implementation of an influential neural image-captioning architecture.