Torch code for Visual Question Answering using a CNN+LSTM model
A Torch implementation of a VIS+LSTM model for answering questions about images using deep learning.
What is Torch code for Visual Question Answering using a CNN+LSTM model?
neural-vqa is a Torch-based deep learning model for Visual Question Answering (VQA), which answers natural language questions about images. It implements the VIS+LSTM architecture from academic research, combining convolutional neural network features with recurrent networks to understand both visual and textual inputs.
Researchers and students in machine learning, computer vision, and NLP who want to experiment with or understand multimodal AI models for visual question answering.
It offers a clean, documented implementation of a published VQA model with pre-trained checkpoints, making it easier to reproduce research results or build upon the architecture for new experiments.
Overview
:grey_question: Visual Question Answering in Torch
Use Cases
Best For
- Reproducing the VIS+LSTM VQA model from the Ren et al. paper
- Learning how to combine CNN image features with LSTM for multimodal tasks
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
