Recurrent Space-time Graph Neural Networks

Authors: Andrei Nicolicioiu, Iulia Duta, Marius Leordeanu

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate, through extensive experiments and ablation studies, that our model outperforms strong baselines and top published methods on recognizing complex activities in video. Moreover, we obtain state-of-the-art performance on the challenging Something-Something human-object interaction dataset. 3 Experiments We perform experiments on two video classification tasks, which involve complex object interactions.
Researcher Affiliation Collaboration Andrei Nicolicioiu , Iulia Duta Bitdefender, Romania anicolicioiu, iduta@bitdefender.com Marius Leordeanu Bitdefender, Romania Institute of Mathematics of the Romanian Academy University "Politehnica" of Bucharest marius.leordeanu@imar.ro
Pseudocode Yes Algorithm 1 Space-time processing in RSTG model.
Open Source Code Yes The code for the full model can be found in our repository 2. https://github.com/Iulia Duta/RSTG
Open Datasets Yes We experiment on a video dataset that we create synthetically, containing complex patterns of movements and shapes, and on the challenging Something-Something-v1 dataset, involving interactions between a human and other objects [54]. [54] Raghav Goyal, Samira Ebrahimi Kahou, Vincent Michalski, Joanna Materzynska, Susanne Westphal, Heuna Kim, Valentin Haenel, Ingo Fruend, Peter Yianilos, Moritz Mueller-Freitag, et al. The" something something" video database for learning and evaluating visual common sense. In ICCV, volume 1, page 3, 2017.
Dataset Splits Yes It consists of a collection of 108499 videos with 86017, 11522 and 10960 videos for train, validation and test splits respectively.
Hardware Specification Yes We show the compute times for different variants of our model and for the Non-Local model using the Resnet-50 backbone on Something-Something videos running on one Nvidia GTX 1080 Ti GPU in Figure 4.
Software Dependencies No We implement our model in Tensorflow framework [58]. The paper mentions TensorFlow but does not provide specific version numbers for it or any other software dependencies.
Experiment Setup Yes We use cross-entropy as loss function and trained the model end-to-end with SGD with Nesterov Momentum with value 0.9 for momentum, starting from a learning rate of 0.0001 and decreasing by a factor of 10 when performance saturates.