Phase-based frame interpolation for video

By Simone Schaub-Meyer, Oliver Wang, Henning Zimmer, Max Grosse, Alexander Sorkine-Hornung. 2015

302 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric (2018, methodology): “Frame interpolation We sample patches from different frame interpolation algorithms, including three variants of flow-based interpolation [33], CNN-based interpolation [39], and phase-based interpolation [37] on the Davis Middleburry dataset [49].”
Video Frame Interpolation via Adaptive Separable Convolution (2017, methodology): “Different from optical ﬂow based methods, Meyer et al . developed a phase-based interpolation method that represents motion in the phase shift of individual pixels and generates intermediate frames by per-pixel phase modiﬁcation [35].”
MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement (2018, methodology): “terpolated frames. The outputs are usually blurry and contain fewer details as this deep model is not able to capture the multi-modal distribution of natural images and videos. The phase-based method [33] manipulates the pixel phase information within a multi-scale pyramid for frame interpolation. H”
Video Frame Interpolation via Adaptive Convolution (2017, methodology): “Following recent frame interpolation work [36], we use the interpolation method from the Middlebury benchmark [1] to synthesize the interpolated frame using the optical flow results.”
Light field video capture using a learning-based hybrid imaging system (2017, unknown): “Many researchers have tried to upsample the frame rate of 2D videos by interpolating frames in the temporal domain (Baker et al. 2011; Liao et al. 2014; Mahajan et al. 2009; Meyer et al. 2015).”
Light Field Image Processing: An Overview (2017, background): “Most researchers have focused on video frame interpolation [91], [92].”
Softmax Splatting for Video Frame Interpolation (2020, background): “Research on frame interpolation has seen a recent resurgence, with multiple papers proposing kernel-based [3, 4, 38, 39], ﬂow-based [3, 4, 22, 28, 29, 37, 41, 43, 51], and phase-based [34, 35] approaches.”
Time-Agnostic Prediction: Predicting Predictable Video Frames (2018, background): “In addition to future prediction, we also consider bidirectional prediction, studied before mainly in the context of video interpolation with very short timescales (< 1s) [21, 25].”
Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (2017, methodology): “2 input images actual in-between PhaseBased [18] FlowNet2 [1, 9] DVF [15] SepConv [20] ours Frame Index It can be clearly seen that our approach produces the best predictions for each individual in-between time step except slightly worse than SepConv at the last time step.”
Channel Attention Is All You Need for Video Frame Interpolation (2020, methodology): “The phase-based interpolation methods (Meyer et al. 2015; 2018) represent motion with the phase shift of individual pixels and construct intermediate frames using modified per-pixel phase.”
Context-Aware Synthesis for Video Frame Interpolation (2018, methodology): “Since the interpolation category of the Middlebury optical flow benchmark is typically used for assessing frame interpolation methods [1, 31, 33, 34], we compare our approach with the methods that perform best on this interpolation benchmark and that are publicly available.”
Hierarchical Style-based Networks for Motion Synthesis (2020, methodology): “Previous researches often utilize phase dynamics [21], flow based feature [20] and other motion information [23,40,38] to facilitate this task.”
Real-Time Intermediate Flow Estimation for Video Frame Interpolation (2020, background): “Meyer et al. [38] utilize phase information to learn the motion relationship for multiple video frame interpolation.”
Video Frame Synthesis Using Deep Voxel Flow (2017, methodology): “[22] employ a Eulerian, phase-based approach to interpolation, but the method is limited to smaller motions.”
Deep Video Deblurring for Hand-Held Cameras (2017, result): “Multi-scale phase-based methods have proven to be able to generate sharp images using purely Eulerian representations [31], so we experimented with multiscale-supervised, Laplacian reconstructions, but found similarly inconclusive results.”
Video Acceleration Magnification (2017, unknown): “Examples include phase-based video frame interpolation [18] and video modification transfer [17].”