1-83 of about 83 matches for site:arxiv.org motions
https://arxiv.org/abs/2312.04966
2312.04966] NewMove: Customizing text-to-video models with novel motions Skip to main content We gratefully
https://arxiv.org/abs/2504.15376
2504.15376] Towards Understanding Camera Motions in Any Video Skip to main content We gratefully acknowledge support
https://arxiv.org/abs/1203.4219
1203.4219] Evidence of Galaxy Cluster Motions with the Kinematic Sunyaev-Zel'dovich Effect Skip to
https://arxiv.org/abs/2409.18127
2409.18127] EgoLM: Multi-Modal Language Model of Egocentric Motions Happy Open Access Week from arXiv! YOU make open
http://arxiv.org/abs/1203.4219
1203.4219] Evidence of Galaxy Cluster Motions with the Kinematic Sunyaev-Zel'dovich Effect Skip to
https://arxiv.org/abs/2412.13196
enabling real-world humanoid robots to perform expressive and dynamic whole-body motions while maintaining overall stability
https://arxiv.org/abs/2412.13196
enabling real-world humanoid robots to perform expressive and dynamic whole-body motions while maintaining overall stability
[2311.07446] Story-to-Motion: Synthesizing Infinite and Controllable Character Animation from Long T
https://arxiv.org/abs/2311.07446
are required to move to various locations and perform specific motions based on a
https://arxiv.org/abs/2404.01203
end frame. In order to achieve high fidelity and generate motions unseen in the
https://arxiv.org/abs/2505.04999
find that existing methods struggle when applied to complex robot tasks requiring fine-grained motions. We design continuous latent
https://arxiv.org/abs/2310.08580
complementary for balancing control accuracy and motion realism. By combining them, OmniControl generates motions that are realistic, coherent
https://arxiv.org/abs/2502.20370
Previously, two main settings existed for two-character interaction generation: (1) generating one's motions based on the
https://arxiv.org/abs/2311.18303
animal domain. We jointly train motion autoencoders for both animal and human motions and at
https://arxiv.org/abs/2311.18303
animal domain. We jointly train motion autoencoders for both animal and human motions and at
https://arxiv.org/abs/2405.18418
policies in 8 tasks with a simulated 56-DoF humanoid, while synthesizing motions that are broadly preferred
https://arxiv.org/abs/1409.6733
line of sight. However, beyond the solar neighborhood, the corresponding proper motions have generally been too
https://arxiv.org/abs/2312.04547
capable of engaging in social interactions and expressing with articulated body motions, thereby simulating life in
https://arxiv.org/abs/2312.04547
capable of engaging in social interactions and expressing with articulated body motions, thereby simulating life in
https://arxiv.org/abs/2206.14797
method learns a rich embedding of decomposable 3D structures and motions that enables new visual
https://arxiv.org/abs/2506.14770
View PDF HTML (experimental) Abstract: The ability to track general whole-body motions in the
https://arxiv.org/abs/2506.14770
View PDF HTML (experimental) Abstract: The ability to track general whole-body motions in the
https://arxiv.org/abs/2405.07784
by Zhi Cen and 7 other authors View PDF HTML (experimental) Abstract: Generating human motions from textual descriptions has
https://arxiv.org/abs/2403.04436
scalable "sim-to-data" process to filter and pick feasible motions using a privileged
https://arxiv.org/abs/2405.07784
by Zhi Cen and 7 other authors View PDF HTML (experimental) Abstract: Generating human motions from textual descriptions has
https://arxiv.org/abs/2403.12959
key observations. Firstly, camera-frame SMPL-X estimation methods readily recover absolute human depth. Secondly, human motions inherently provide absolute spatial
https://arxiv.org/abs/2312.17135
InsActor to capture complex relationships between high-level human instructions and character motions by employing diffusion policies
https://arxiv.org/abs/2012.09855
capabilities of current view synthesis methods, which quickly degenerate when presented with large camera motions. Methods for video
https://arxiv.org/abs/2104.08381
encourages the object detector to explore invariant structures across instances under various motions, which leads to
https://arxiv.org/abs/2303.05703
factorizing the scene motion as a composition of part-level rigid motions. Experimentally, our method can
http://arxiv.org/abs/1402.0661
525 have motions that can be independently confirmed from earlier 2MASS images yet lack any published motions in SIMBAD
https://arxiv.org/abs/2310.08580
complementary for balancing control accuracy and motion realism. By combining them, OmniControl generates motions that are realistic, coherent
[2203.09457] Look Outside the Room: Synthesizing A Consistent Long-Term 3D Scene Video from A Single
https://arxiv.org/abs/2203.09457
rendering techniques. However, most work is still limited by synthesizing new views within relatively small camera motions. In this
https://arxiv.org/abs/2201.05124
here we report an analysis of the 3D positions, shapes, and motions of dense
[2502.01143] ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body
https://arxiv.org/abs/2502.01143
in performing human-like, whole-body skills. However, achieving agile and coordinated whole-body motions remains a significant
https://arxiv.org/abs/2402.16796
Abstract: Can we enable humanoid robots to generate rich, diverse, and expressive motions in the
https://arxiv.org/abs/2011.15119
RL), which learn control from (human) demonstrations. While RL has shown impressive results at reproducing individual motions and interactive
https://arxiv.org/abs/2412.07755
space, i.e., reasoning about the effect of egocentric and object motions on spatial relationships. Manually
https://arxiv.org/abs/2412.07773
Predictive Motion Priors), trained with Conditional Variational Autoencoder (CVAE) to effectively represent upper-body motions. The locomotion
http://arxiv.org/abs/2104.08381
encourages the object detector to explore invariant structures across instances under various motions, which leads to
https://arxiv.org/abs/2502.12152
smoothness or speed / torque limits. The second stage then refines the discovered motions into deployable (i.e. smooth
https://arxiv.org/abs/2412.07773
Predictive Motion Priors), trained with Conditional Variational Autoencoder (CVAE) to effectively represent upper-body motions. The locomotion
https://arxiv.org/abs/2011.15119
RL), which learn control from (human) demonstrations. While RL has shown impressive results at reproducing individual motions and interactive
https://arxiv.org/abs/2410.18912
learning the neural dynamics model on offline robot interaction data, our method can predict object motions under varying initial configurations
https://arxiv.org/abs/2007.11678
a physically-plausible motion, based on the inputs. We show this process produces motions that are significantly more
https://arxiv.org/abs/2109.09913
contacts in a differentiable way. This optimization yields corrected 3D poses and motions, as well as their
https://arxiv.org/abs/2308.11617
temporal consistency in the latent space (LTC), and generate consistent interaction motions. In the
https://arxiv.org/abs/2410.18912
learning the neural dynamics model on offline robot interaction data, our method can predict object motions under varying initial configurations
https://arxiv.org/abs/2403.18074
tandem, our method learns a latent that encodes representations of general repetitive motions, which we use for
https://arxiv.org/abs/2211.11082
have shown impressive results on this task. However, for long videos with complex object motions and uncontrolled
https://arxiv.org/abs/2106.04004
and 6 other authors View PDF Abstract: A deep generative model that describes human motions can benefit a
https://arxiv.org/abs/2204.09739
allows for the detection of small separation binaries with significant proper motions. We used the
[2208.01160] Hierarchical Reinforcement Learning for Precise Soccer Shooting Skills using a Quadrupe
https://arxiv.org/abs/2208.01160
to train (a) a robust motion control policy that can track arbitrary motions and (b
https://arxiv.org/abs/2205.08535
and texture, and drive the avatar with the described motions using solely natural languages
https://arxiv.org/abs/2202.10448
human to control a robot hand and arm, simply by demonstrating motions with their own hand
https://arxiv.org/abs/2204.00604
conditioned on dance videos. Our proposed framework takes dance video frames and human body motions as input, and
https://arxiv.org/abs/2405.17405
of previous methods based on generative adversarial networks or vanilla diffusion models, which struggle with complex motions, viewpoint changes, and
[2312.06762] "Beads on a String" Star Formation Tied to one of the most Powerful AGN Outbursts Obser
https://arxiv.org/abs/2312.06762
the young stars and the molecular gas to turbulent intracluster gas motions, and suggest
[2312.08983] Interactive Humanoid: Online Full-Body Motion Reaction Synthesis with Social Affordance
https://arxiv.org/abs/2312.08983
body motion reaction synthesis, which generates humanoid reactions based on the human actor's motions. The previous
https://arxiv.org/abs/2404.04421
quality novel-view renderings of avatars dressed in loose-fitting clothes under motions and lighting
https://arxiv.org/abs/2007.11678
a physically-plausible motion, based on the inputs. We show this process produces motions that are significantly more
https://arxiv.org/abs/2308.11617
temporal consistency in the latent space (LTC), and generate consistent interaction motions. In the
https://arxiv.org/abs/2109.09913
contacts in a differentiable way. This optimization yields corrected 3D poses and motions, as well as their
https://arxiv.org/abs/2202.10448
human to control a robot hand and arm, simply by demonstrating motions with their own hand
https://arxiv.org/abs/2204.00604
conditioned on dance videos. Our proposed framework takes dance video frames and human body motions as input, and
https://arxiv.org/abs/2205.08535
and texture, and drive the avatar with the described motions using solely natural languages
https://arxiv.org/abs/2405.17405
of previous methods based on generative adversarial networks or vanilla diffusion models, which struggle with complex motions, viewpoint changes, and
https://arxiv.org/abs/2404.10667
producing a large spectrum of facial nuances and natural head motions that contribute to
https://arxiv.org/abs/2210.10044
arm and legs, and error can propagate across modules causing non-smooth unnatural motions. It is also biological
https://arxiv.org/abs/2404.10667
producing a large spectrum of facial nuances and natural head motions that contribute to
[1509.06373] The clustering of galaxies in the SDSS-III Baryon Oscillation Spectroscopic Survey: BAO
http://arxiv.org/abs/1509.06373
applying a reconstruction algorithm to reduce the BAO suppression by bulk motions, we measure the
http://arxiv.org/abs/physics/0310081
density of mass is assumed uniformly distributed along an infinitely long straight line. All possible motions of a
https://arxiv.org/abs/2404.04421
quality novel-view renderings of avatars dressed in loose-fitting clothes under motions and lighting
https://arxiv.org/abs/2310.16035
show that LEFT flexibly learns concepts in four domains: 2D images, 3D scenes, human motions, and robotic
https://arxiv.org/abs/2304.12317
from diverse viewpoints (which helps reconstruction) but are also more likely to contain larger motions (which complicates reconstruction). To
https://arxiv.org/abs/2404.12391
sampling from a large set of generated videos that do not contain motions, one can drastically decrease
https://arxiv.org/abs/2304.12317
from diverse viewpoints (which helps reconstruction) but are also more likely to contain larger motions (which complicates reconstruction). To
https://arxiv.org/abs/2404.12391
sampling from a large set of generated videos that do not contain motions, one can drastically decrease
http://arxiv.org/archive/bayes-an/
Classical Physics ( new , recent , current month ) Newtonian and relativistic dynamics; many particle systems; planetary motions; chaos in classical
https://arxiv.org/abs/2306.17749
useful for cosmological and astrophysical quasar studies. We apply cuts based on proper motions and Gaia
[2008.06396] Spitzer Follow-up of Extremely Cold Brown Dwarfs Discovered by the Backyard Worlds: Pla
https://arxiv.org/abs/2008.06396
of our discoveries. Nine of our motion-confirmed objects have best-fit linear motions larger than 1"/yr
[2008.06396] Spitzer Follow-up of Extremely Cold Brown Dwarfs Discovered by the Backyard Worlds: Pla
https://arxiv.org/abs/2008.06396
of our discoveries. Nine of our motion-confirmed objects have best-fit linear motions larger than 1"/yr