Patch to the Future: Unsupervised Visual Prediction

By Jacob Walker, Abhinav Kumar Gupta, Martial Hebert. 2014

207 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Generative Models

Citations in the corpus

Generating Videos with Scene Dynamics (2016, background): “Our work is also related to efforts to predict the future in video [33, 22, 44, 51, 42, 17, 8, 55] as well as concurrent work in future generation [6, 15, 20, 50, 56, 43].”
Contrastive Bidirectional Transformer for Temporal Representation Learning (2019, background): “The capacity of reasoning over longtemporal horizons is critical for many applications, such as video storytelling [41, 29] and action anticipation [32].”
Unsupervised Learning for Physical Interaction through Video Prediction (2016, background): “Learning physics: Several works have explicitly addressed prediction of physical interactions, including predicting ball motion [5], block falling [2], the effects of forces [19, 18], future human interactions [9], and future car trajectories [26].”
An Uncertain Future: Forecasting from Static Images Using Variational Autoencoders (2016, background): “One way this limitation has been addressed is by breaking the images into smaller pieces based on mid-level discriminative patches [3].”
"What Happens If..." Learning to Predict the Effect of Forces in Images (2016, background): “[36] infers the future path of rigid objects according to learned models of appearance, context, and transition.”
Newtonian Image Understanding: Unfolding the Dynamics of Objects in Static Images (2015, background): “[33] proposed to infer the most likely path for objects.”
Learning to Generate Long-term Future via Hierarchical Prediction (2017, background): “Other existing works focus on predicting high-level semantics, such as trajectories or action labels (Walker et al., 2014; Chao et al., 2017; Yuen & Torralba, 2010; Lee, 2015), driven by immediate applications (e.g., video surveillance).”
Physics 101: Learning Physical Object Properties from Unlabeled Videos (2016, methodology): “Our incorporation of physical laws into our model distinguish our framework from previous methods introduced in the computer vision and robotics community for predicting physical interactions or properties of objects for various purposes [4, 5, 9, 12, 16, 21, 22, 26].”
Dense Optical Flow Prediction from a Static Image (2015, background): “The most common approach to this prediction problem is to use a planning-based agent-centric approach: an object [10] or a patch [24] is modeled as an agent that performs actions based on its current state and the goal state.”
Learning Temporal Transformations from Time-Lapse Videos (2016, methodology): “Future prediction: Future prediction has been applied to various tasks such as estimating the future trajectories of cars[11], pedestrians[12], or general objects[13] in images or videos.”
Inferring the Why in Images (2014, background): “There also has been work in forecasting activities [16, 32] and early event detection [11], but they are interested in predicting the future in videos while we wish to explain the motivations of actions.”
Anticipating the future by watching unlabeled video (2015, background): “[45] demonstrate a compelling data-driven approach that animates the trajectory of objects from a single frame.”
Predicting Motivations of Actions by Leveraging Text (2014, background): “There also has been work in forecasting activities [18, 37], inferring goals [40], and detecting early events [12], but they are interested in predicting the future in videos while we wish to explain the motivations of actions of people in images.”
Forecasting Interactive Dynamics of Pedestrians with Fictitious Play (2016, background): “Patterns of avoidance can also be used to learn the hidden rewards or costs of physical spaces [21, 44, 41].”
Learning Video Representations using Contrastive Bidirectional Transformer (2019, background): “sentations for both the semantics and the dynamics. The capacity of reasoning over longtemporal horizons is critical for many applications, such as video storytelling [41, 29] and action anticipation [32]. This paper aims at learning representations from long videos. We assume that this task can be ”
Anticipating Visual Representations from Unlabeled Video (2015, background): “Building on the rich potential of large video collections, Walker et al. [39] demonstrate a compelling data-driven approach that animates the trajectory of objects from a single frame.”
Human motion trajectory prediction: a survey (2019, background): “2019), implicitly as a result of pixel-wise prediction (Walker et al. 2014), or by learning a joint motion policy (Ma et al.”
Human Action Recognition and Prediction: A Survey (2018, background): “in [193] predicted the behavior of agents (e.”