In Defense of Image Pre-Training for Spatiotemporal Recognition. (arXiv:2205.01721v1 [cs.CV])

Web: http://arxiv.org/abs/2205.01721

Image pre-training, the current de-facto paradigm for a wide range of visual
tasks, is generally less favored in the field of video recognition. By
contrast, a common strategy is to directly train with spatiotemporal
convolutional neural networks (CNNs) from scratch. Nonetheless, interestingly,
by taking a closer look at these from-scratch learned CNNs, we note there exist
certain 3D kernels that exhibit much stronger appearance modeling ability than
others, arguably suggesting appearance information is already well disentangled
in learning. Inspired by …

arxiv cv defense image pre-training training