AViD Dataset: Anonymized Videos from Diverse Countries

10 Jul 2020AJ PiergiovanniMichael S. Ryoo

We introduce a new public video dataset for action recognition: Anonymized Videos from Diverse countries (AViD). Unlike existing public video datasets, AViD is a collection of action videos from many different countries... (read more)

PDF Abstract

Results from the Paper


 Ranked #1 on Action Detection on Charades (using extra training data)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
RESULT BENCHMARK
Action Classification AViD (2+1)D ResNet-50 Accuracy 48.8 # 5
Action Classification AViD SlowFast-50 4x4 Accuracy 48.5 # 6
Action Classification AViD SlowFast-50 8x8 Accuracy 50.4 # 3
Action Classification AViD 2D ResNet-50 Accuracy 36.2 # 9
Action Classification AViD I3D Accuracy 46.8 # 8
Action Classification AViD 3D ResNet-50 Accuracy 48.2 # 7
Action Classification AViD Two-Stream 3D ResNet-50 Accuracy 50.1 # 4
Action Classification AViD RepFlow ResNet-50 Accuracy 50.5 # 2
Action Classification AViD SlowFast-101 16x8 Accuracy 50.9 # 1
Action Detection Charades 3D ResNet-50 pretrained on AViD mAP 23.2 # 2
Action Detection Charades 3D ResNet-50 + super-events pretrained on AViD mAP 25.2 # 1

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet