ConvNet Architecture Search for Spatiotemporal Feature Learning
read the original abstract
Learning image representations with ConvNets by pre-training on ImageNet has proven useful across many visual understanding tasks including object detection, semantic segmentation, and image captioning. Although any image representation can be applied to video frames, a dedicated spatiotemporal representation is still vital in order to incorporate motion patterns that cannot be captured by appearance based models alone. This paper presents an empirical ConvNet architecture search for spatiotemporal feature learning, culminating in a deep 3-dimensional (3D) Residual ConvNet. Our proposed architecture outperforms C3D by a good margin on Sports-1M, UCF101, HMDB51, THUMOS14, and ASLAN while being 2 times faster at inference time, 2 times smaller in model size, and having a more compact representation.
This paper has not been read by Pith yet.
Forward citations
Cited by 2 Pith papers
-
BMN: Boundary-Matching Network for Temporal Action Proposal Generation
BMN uses a boundary-matching mechanism to generate precise temporal action proposals with reliable confidence scores simultaneously in an end-to-end framework, improving results on THUMOS-14 and ActivityNet-1.3.
-
AVD: Adversarial Video Distillation
AVD maps videos to semantically realistic 2D images via 3D conv encoder-decoder plus adversarial training, enabling image-based classifiers to perform video activity recognition.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.