First-person hyper-lapse videos

By Johannes Peter Kopf, Michael F. Cohen, Richard Szeliski. 2014

199 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Deep Stereo: Learning to Predict New Views from the World's Imagery (2015, neutral): “This can be used for applications such as cinematography, virtual reality, teleconferencing [4], image stabilization [19], or 3-dimensionalizing monocular ﬁlm footage.”
Structure from motion photogrammetry in physical geography (2016, unknown): “Video stabilisation algorithms, such as those underlying the ‘hyper-lapse’ videos of Kopf et al. (2014), could improve the quality of video footage taken onboard mobile platforms.”
BARF: Bundle-Adjusting Neural Radiance Fields (2021, background): “Researchers have investigated blending pixel colors based on depth maps [4] or leveraging proxy geometry to warp and composite the synthesized image [22].”
Video summarization by learning submodular mixtures of objectives (2015, background): “Videos can be summarized into many different representations: Keyframes [37] [15] [11] [12], skims [9] [24], storyboards [4], time-lapses [13], montages [34] or video synopses [30].”
Enhancing Photorealism Enhancement (2021, background): “real imagery of a scene for rendering novel views [36], [37], [38], [39], [40], [41], [42], [43], [44].”
First Person Action-Object Detection with EgoNet (2016, methodology): “ual attention and motor actions [10]. This ﬁrst person property has been exploited in various tasks such as summarizing daily activities using a ﬁrst person video [11,12], creating a hyperlapse video [13], detecting objects [14,15], and recognizing activities [16–20]. In particular, egocentric spati”
Deep blending for free-viewpoint image-based rendering (2018, methodology): “2001], the more sophisticated texture-stretch heuristics of Hyperlapse [Kopf et al. 2014b] ś requiring an oline graphcut/Poisson blending step ś or the adaptive bandwidth of InsideOut [Hedman et al.”
Summarization of Egocentric Videos: A Comprehensive Survey (2017, methodology): “7) Microsoft’s sports dataset [27]: Used for fast-forwarding objectives, it was recorded with a GoPro camera on a”
Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis (2023, background): “Early approaches leverage image-based rendering techniques with proxy geometry/depth to sample novel views from source images [11, 13, 20, 31, 34, 44, 46, 103].”
ACNe: Attentive Context Normalization for Robust Permutation-Equivariant Learning (2020, background): “It lies at the core of Structurefrom-Motion (SfM), which, in turn, is the building block of applications such as 3D reconstruction [1], image-based rendering [43] and time-lapse smoothing [28].”