The region feature file should be decompressed and placed under your feature directory. The region features and detections are available for download ( feature and detection). Then, download the ground-truth caption annotations (under our val/test splits) from here and same place under data/anet.
Or you can reproduce them all using the data from ActivityNet-Entities repo and the preprocessing script prepro_dic_anet.py under prepro. The following files have been updated to include the hidden test set or video IDs: anet_detection_vg_fc6_feat_100rois.h5, anet_entities_, and anet_entities_.ĭownload the preprocessed annotation files from here, uncompress and place them under data/anet. Make sure you move the additional *.npy files over to your folder fc6_feat_100rois and rgb_motion_1d, respectively. Updates on : Feature files for the hidden test set, used in ANet-Entities Object Localization Challenge 2020, are available to download ( region features and frame-wise features). Download Stanford CoreNLP 3.9.1 for grounding evaluation and place the uncompressed folder under the tools directory. Also, download and place the reference file under coco-caption/annotations. (Optional) If you choose to not use download_all.sh, be sure to install JAVA and download Stanford CoreNLP for SPICE (see here).Replace cfgs/conda_env_gvd_p圓.yml with cfgs/conda_env_gvd.yml for Python 2.7. This code base could potentially work with PyTorch 1.2+ with corresponding changes made.
Note that there have been some breaking changes since PyTorch 1.2 (e.g., bitwise not on torch.bool/torch.uint8 and masked_fill_). MINICONDA_ROOT=Ĭonda env create -f cfgs/conda_env_gvd_p圓.yml -prefix $MINICONDA_ROOT/envs/gvd_pytorch1.1
0 kommentar(er)
