Deep View Morphing

Ji, Dinghuang; Junghyun, Kwon; McFarland, Max; Savarese, Silvio

doi:10.48550/arxiv.1703.02168

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With a huge amount of multiview images, 3D stereo algorithms [5] are applicable to reconstruct the 3D scene and then be utilized to synthesize novel views. Ji et al [11] proposed to synthesize middle view images by using two rectified view images. Yan et al [33] proposed a perspective transformer network to learn the projection transformation after reconstructing the 3D volume of the object.…”

Section: Related Workmentioning

confidence: 99%

“…View synthesis is a long-standing problem in computer vision [5,11,25,33,35], which facilitates many applications including surrounding perception and virtual reality. In modern autonomous driving solution, the limited viewpoint of on-car cameras restricts the system from reliably understanding the environment, acquiring accurate global view for better policy making and path planning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Generative Adversarial Frontal View to Bird View Synthesis

Zhu

Yin²,

Shi³

et al. 2018

2018 International Conference on 3D Vision (3DV)

View full text Add to dashboard Cite

Environment perception is an important task with great practical value and bird view is an essential part for creating panoramas of surrounding environment. Due to the large gap and severe deformation between the frontal view and bird view, generating a bird view image from a single frontal view is challenging. To tackle this problem, we propose the BridgeGAN, i.e., a novel generative model for bird view synthesis. First, an intermediate view, i.e., homography view, is introduced to bridge the large gap. Next, conditioned on the three views (frontal view, homography view and bird view) in our task, a multi-GAN based model is proposed to learn the challenging cross-view translation. Extensive experiments conducted on a synthetic dataset have demonstrated that the images generated by our model are much better than those generated by existing methods, with more consistent global appearance and sharper details. Ablation studies and discussions show its reliability and robustness in some challenging cases. Codes are available at

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Generative Adversarial Frontal View to Bird View Synthesis

Zhu

Yin²,

Shi³

et al. 2018

2018 International Conference on 3D Vision (3DV)

View full text Add to dashboard Cite

show abstract

“…One work focused on developing generative models for images in random poses, without regard to correspondence to some input sequence of poses [1]. Another focuses on interpolating between different views of an object by generating dense correspondences between pixels [6]. A final paper discussed using a two-stage process for synthesizing person images in arbitrary poses, generating a coarse structure before refining it [8].…”

Section: Related Workmentioning

confidence: 99%

Generative Models for Pose Transfer

Chao,

Li,

Swamy

2018

Preprint

View full text Add to dashboard Cite

We investigate nearest neighbor and generative models for transferring pose between persons. We take in a video of one person performing a sequence of actions and attempt to generate a video of another person performing the same actions. Our generative model (pix2pix) outperforms k-NN at both generating corresponding frames and generalizing outside the demonstrated action set. Our most salient contribution is determining a pipeline (pose detection, face detection, k-NN based pairing) that is effective at performing the desired task. We also detail several iterative improvements and failure modes.

show abstract

Deep View Morphing

Cited by 2 publications

References 0 publications

Generative Adversarial Frontal View to Bird View Synthesis

Generative Adversarial Frontal View to Bird View Synthesis

Generative Models for Pose Transfer

Contact Info

Product

Resources

About