Future Semantic Segmentation with Convolutional LSTM
arXiv:1807.07946
Abstract
We consider the problem of predicting semantic segmentation of future frames in a video. Given several observed frames in a video, our goal is to predict the semantic segmentation map of future frames that are not yet observed. A reliable solution to this problem is useful in many applications that require real-time decision making, such as autonomous driving. We propose a novel model that uses convolutional LSTM (ConvLSTM) to encode the spatiotemporal information of observed frames for future prediction. We also extend our model to use bidirectional ConvLSTM to capture temporal information in both directions. Our proposed approach outperforms other state-of-the-art methods on the benchmark dataset.
arXiv commentAccepted to BMVC 2018
References in corpus (8)
- Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- Decomposing Motion and Content for Natural Video Sequence Prediction
- Video (language) modeling: a baseline for generative models of natural videos
- Unsupervised Learning for Physical Interaction through Video Prediction
- Learning to Generate Long-term Future via Hierarchical Prediction
- Semantic Video Segmentation by Gated Recurrent Flow Propagation
- Low-Latency Video Semantic Segmentation
Cited by in corpus (8)
- A Review on Deep Learning Techniques for Video Prediction
- Frame-To-Frame Consistent Semantic Segmentation
- Multi-Modal Temporal Convolutional Network for Anticipating Actions in Egocentric Videos
- Noisy-LSTM: Improving Temporal Awareness for Video Semantic Segmentation
- A Study on Trees's Knots Prediction from their Bark Outer-Shape
- A Survey On 3D Inner Structure Prediction from its Outer Shape
- An Abstraction Model for Semantic Segmentation Algorithms
- Video Region Annotation with Sparse Bounding Boxes