PredRNN: Recurrent Neural Networks for Predictive Learning using Spatiotemporal LSTMs

By Yunbo Wang, Mingsheng Long, Jianmin Wang, Zhifeng Gao, Philip S. Yu. Neural Information Processing Systems, 2017

1,007 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/566d9637c465e4dfc1c1695ecc7155adbb54f45c

How researchers see this paper

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

Citations in the corpus

Skilful precipitation nowcasting using deep generative models of radar (2021, methodology): “On that same resized dataset, an alternative spatio-temporal architecture with stacked RNN layers was proposed [12].”
Self-Attention ConvLSTM for Spatiotemporal Prediction (2020, methodology): “ConvLSTM (Shi et al. 2015) replaces all the linear operations in it with convolution layers to capture spatial dependencies besides the long-short term modeling, and many of its variants (Wang et al. 2017a; 2018b; 2019) have achieved impressive results on spatiotemporal prediction.”
Deep Learning for Spatio-Temporal Data Mining: A Survey (2019, unknown): “GraphCNN [78], [146], [86], [102], [135], [22], [84], [164], [42], [8], [78], [146] RNN(LSTM,GRU) [40], [70], [156], [90], [83], [154], [34], [150], [57], [37], [126], [169], [80], [74], [178], [36], [160], [157], [39], [58], [180] [117], [26], [166], [82], [23], [81], [167], [17], [168], [92], [88]”
PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning (2021, methodology): “Building upon the first version of ST-LSTM [2], we present a new decoupling loss, as shown in Fig.”
Small Sample Learning in Big Data Era (2018, methodology): “The former manner tends to generate sample redundancy and encounters the problem of over-fitting to the rare classes, whereas the latter inclines to lose critical feature knowledge within the classes with more samples (Wang et al., 2017c).”
Unsupervised Learning From Video With Deep Neural Embeddings (2019, background): “Temporal autoencoders such as PredNet [22], PredRNN [35], and PredRNN++ [34] are intriguing but have not yet evidenced substantial transfer learning performance at scale.”
Can deep learning beat numerical weather prediction? (2021, methodology): “Based on the concept of video prediction, various types of networks were used for short-term prediction of sky images and radar images [46,79,81].”
SmaAt-UNet: Precipitation Nowcasting using a Small Attention-UNet Architecture (2020, background): “Recent advances in artiﬁcial neural network architectures (ANNs) have enabled data-driven based models to bridge the present gap for short-term forecasting [6, 7, 8, 9].”
A review of convolutional neural networks in computer vision (2024, unknown): “Wang et al. (2019) presented a three-dimensional CNN architecture in conjunction with LSTM to distinguish various activities from video frames in order to address the issues of poor dynamic information obtained by the model, low prediction accuracy, and bad quality of the produced pictures.”
Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Survey (2023, background): “These hybrid networks (e.g., ConvLSTM [9], PredRNN [10]) have been increasingly applied to predictive learning of urban spatio-temporal data and have shown signiﬁcant advantages.”
SimVP: Simpler yet Better Video Prediction (2022, methodology): “Sixteen baselines are included, such as MCnet [62], ConvLSTM [71], SAVP, SAVP-VAE [30], VPN [28], DFN [25], fRNN [43], Znet [74], SV2P [3], PredRNN [67], VarNet [27], PredRNN++ [65], MSNET [31], E3d-LSTM [66] and STMFANet [26].”
CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers (2022, methodology): “Early works, e.g. CDNA [8] and PredRNN [32], leverage deterministic methods to directly predict the next frame via CNNs or RNNs.”
GODIVA: Generating Open-DomaIn Videos from nAtural Descriptions (2021, background): “Further, [29] proposed stacked ConvLSTM, which shares the hidden state among the layers in the stack.”
A Review on Deep Learning Techniques for Video Prediction (2020, methodology): “CrevNet [154] 2020 [13], [196], [197] 3d-cED [74], [119], [120], [198] RGB RGB X MSE SR ? ? ? X X”
Phenaki: Variable Length Video Generation From Open Domain Textual Description (2022, methodology): “These papers include deterministic methods using a combination of recurrent and convolutional networks [36, 42, 13, 50], variational based stochastic meth-ods [2, 10, 46, 3] and more recently by learning a discrete representation [49, 33, 31], auto-regressive models [51, 55, 28, 57], diffusion…”
A review on the long short-term memory model (2020, background): “In Wang et al. (2017) spatio-temporal LSTM was introduced in order to predict the next images based on historical frames which did achieve state-of-the-art prediction performance.”
PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning (2018, methodology): “Due to a signiﬁcant increase in the recurrence depth along the spatiotemporal transition pathway, this newly designed cascaded memory is superior to the simple concatenation structure of the spatiotemporal LSTM (Wang et al., 2017).”