This dataset contains precomputed MS-COCO and Flickr30K Faster R-CNN image features, which are all the data needed for reproducing the experiments in "Stacked Cross Attention for Image-Text Matching", our ECCV 2018 paper. We use splits produced by Andrej Karpathy.

Dataset Files

You must be an IEEE Dataport Subscriber to access these files. Subscribe now or login.

[1] Kuang-Huei Lee, "SCAN Faster R-CNN Image Features", IEEE Dataport, 2024. [Online]. Available: http://dx.doi.org/10.21227/r7ft-ke86. Accessed: Dec. 26, 2024.
@data{r7ft-ke86-24,
doi = {10.21227/r7ft-ke86},
url = {http://dx.doi.org/10.21227/r7ft-ke86},
author = {Kuang-Huei Lee },
publisher = {IEEE Dataport},
title = {SCAN Faster R-CNN Image Features},
year = {2024} }
TY - DATA
T1 - SCAN Faster R-CNN Image Features
AU - Kuang-Huei Lee
PY - 2024
PB - IEEE Dataport
UR - 10.21227/r7ft-ke86
ER -
Kuang-Huei Lee. (2024). SCAN Faster R-CNN Image Features. IEEE Dataport. http://dx.doi.org/10.21227/r7ft-ke86
Kuang-Huei Lee, 2024. SCAN Faster R-CNN Image Features. Available at: http://dx.doi.org/10.21227/r7ft-ke86.
Kuang-Huei Lee. (2024). "SCAN Faster R-CNN Image Features." Web.
1. Kuang-Huei Lee. SCAN Faster R-CNN Image Features [Internet]. IEEE Dataport; 2024. Available from : http://dx.doi.org/10.21227/r7ft-ke86
Kuang-Huei Lee. "SCAN Faster R-CNN Image Features." doi: 10.21227/r7ft-ke86