Automated search discovers Swish activation f(x) = x * sigmoid(βx) that improves top-1 ImageNet accuracy over ReLU by 0.9% on Mobile NASNet-A and 0.6% on Inception-ResNet-v2.
Learnable pooling with Context Gating for video classification
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
Current methods for video analysis often extract frame-level features using pre-trained convolutional neural networks (CNNs). Such features are then aggregated over time e.g., by simple temporal averaging or more sophisticated recurrent neural networks such as long short-term memory (LSTM) or gated recurrent units (GRU). In this work we revise existing video representations and study alternative methods for temporal aggregation. We first explore clustering-based aggregation layers and propose a two-stream architecture aggregating audio and visual features. We then introduce a learnable non-linear unit, named Context Gating, aiming to model interdependencies among network activations. Our experimental results show the advantage of both improvements for the task of video classification. In particular, we evaluate our method on the large-scale multi-modal Youtube-8M v2 dataset and outperform all other methods in the Youtube 8M Large-Scale Video Understanding challenge.
representative citing papers
FOSNet fuses object and scene features via CNN and uses scene coherence loss to report SOTA accuracies of 60.14% on Places2 and 90.37% on MIT Indoor67.
Baidu-UTS won the EPIC-Kitchens challenge by guiding 3D CNN training with object detection features via a Gated Feature Aggregator to improve noun prediction.
citing papers explorer
-
Searching for Activation Functions
Automated search discovers Swish activation f(x) = x * sigmoid(βx) that improves top-1 ImageNet accuracy over ReLU by 0.9% on Mobile NASNet-A and 0.6% on Inception-ResNet-v2.
-
FOSNet: An End-to-End Trainable Deep Neural Network for Scene Recognition
FOSNet fuses object and scene features via CNN and uses scene coherence loss to report SOTA accuracies of 60.14% on Places2 and 90.37% on MIT Indoor67.
-
Baidu-UTS Submission to the EPIC-Kitchens Action Recognition Challenge 2019
Baidu-UTS won the EPIC-Kitchens challenge by guiding 3D CNN training with object detection features via a Gated Feature Aggregator to improve noun prediction.