CVPR 2020

Transform and Tell: Entity-Aware News Image Captioning (CVPR 2020)

Mar 24, 2021

|

43 views

Alasdair Tran

Multi-Head Attention

Computer Vision

Details

We propose an end-to-end model which generates captions for images embedded in news articles. Our model outperforms the previous state of the art by a factor of four in CIDEr score. Paper: https://arxiv.org/abs/2004.08070 Demo: https://transform-and-tell.ml/ Code: https://github.com/alasdairtran/transform-and-tell Acknowledgement: The background music is from https://www.bensound.com

Category: CVPR 2020

Comments

loading...

Reactions (0) | Note

📝 No reactions yet

Be the first one to share your thoughts!

Reactions(0)

Note

loading...

Recommended

[CVPR 2020 Oral] High-dimensional Convolutional Neural Networks for Geometric Pattern Recognition

Chris Choy | Jul 1, 2020

Mapillary Street-Level Sequences: A Dataset for Lifelong Place Recognition - CVPR 2020 (oral)

SLAMLab | Jul 1, 2020

[CVPR 2020] CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus

Florian Kluger | Jul 1, 2020

[CVPR 2020] Meshlet Priors for 3D Mesh Reconstruction

Orazio Gallo | Jul 1, 2020

Orthogonal Convolutional Neural Networks (CVPR 2020)

Peter Wang BE | Jul 1, 2020