Bags of Spacetime Energies for Dynamic Scene Recognition

59 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

C3D: Generic Features for Video Analysis (2014, methodology): “C3D achieves state-of-the-art performance on both datasets using only a linear SVM with simple averaging of clip features while the second best method [14] uses different complex feature encodings (FV, LLC, and dynamic pooling).”
Temporal Residual Networks for Dynamic Scene Recognition (2017, methodology): “The third best performer, [10], is an ancestor of BoSE and is not considered here.”
Learning Spatio-Temporal Representation with Pseudo-3D Residual Networks (2017, methodology): “In general, CNN-based representations exhibits better accuracy than hand-crafted fea-Table Method Dynamic Scene YUPENN [3] 43.1% 80.7% [5] 77.7% 96.2% C3D [31] 87.7% 98.1% ResNet-152 [7] 93.6% 99.2% P3D ResNet 94.6% 99.5% tures.”
Self-Supervised Visual Feature Learning With Deep Neural Networks: A Survey (2019, background): “Due to its powerful ability of simultaneously extracting both spatial and temporal features from multiple frames, the network achieved state-of-the-art on several video analysis tasks including human action recognition [63], action similarity labeling [64], scene classification [65], and object reco”
Learning Spatiotemporal Features with 3D Convolutional Networks (2014, methodology): “Method [29] [39]([25]) [31] [9] [9] [32] Result 90.”