LogoThread Easy
  • 探索
  • 線程創作
LogoThread Easy

Twitter 線程的一站式夥伴

© 2025 Thread Easy All Rights Reserved.

探索

Newest first — browse tweet threads

Keep on to blur preview images; turn off to show them clearly

another linear attention, but this one sounds interesting

another linear attention, but this one sounds interesting

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

avatar for Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Fri Oct 31 07:41:33
though note the industrial approach, the flag is regularly pre-cut to ensure easy ripping a high-IQ engineering nation, even in retardation burgers would probably try to legit rip apart a whole cloth and injure themselves falling

though note the industrial approach, the flag is regularly pre-cut to ensure easy ripping a high-IQ engineering nation, even in retardation burgers would probably try to legit rip apart a whole cloth and injure themselves falling

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

avatar for Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Fri Oct 31 07:39:10
Paper (pdf): https://t.co/POADj0bzsO Project: https://t.co/SEmM7IIDB5

Paper (pdf): https://t.co/POADj0bzsO Project: https://t.co/SEmM7IIDB5

Your guide to radiance fields | Host of the podcast @ViewDependent | Founder and CEO of https://t.co/5MjtfpwEU3 | discord: https://t.co/lrl64WGvlD

avatar for MrNeRF
MrNeRF
Fri Oct 31 07:38:43
Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding • We propose Motion4D, a model that integrates 2D priors from foundation models into a dynamic 3D Gaussian Splatting representation. This achieves consistent motion and semantic modeling from monocular videos. • We design a two-part iterative optimization framework comprising: - Sequential optimization: updates motion and semantic fields in consecutive stages to maintain local consistency. - Global optimization: jointly refines all attributes to ensure long-term coherence. • We introduce iterative motion refinement using 3D confidence maps and adaptive resampling to enhance dynamic scene reconstruction. Semantic refinement corrects 2D semantic inconsistencies through iterative updates with SAM2. • Our Motion4D significantly outperforms both 2D foundation models and existing 3D methods in tasks such as video object segmentation, point-based tracking, and novel view synthesis.

Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding • We propose Motion4D, a model that integrates 2D priors from foundation models into a dynamic 3D Gaussian Splatting representation. This achieves consistent motion and semantic modeling from monocular videos. • We design a two-part iterative optimization framework comprising: - Sequential optimization: updates motion and semantic fields in consecutive stages to maintain local consistency. - Global optimization: jointly refines all attributes to ensure long-term coherence. • We introduce iterative motion refinement using 3D confidence maps and adaptive resampling to enhance dynamic scene reconstruction. Semantic refinement corrects 2D semantic inconsistencies through iterative updates with SAM2. • Our Motion4D significantly outperforms both 2D foundation models and existing 3D methods in tasks such as video object segmentation, point-based tracking, and novel view synthesis.

Paper (pdf): https://t.co/POADj0bzsO Project: https://t.co/SEmM7IIDB5

avatar for MrNeRF
MrNeRF
Fri Oct 31 07:38:37
RT @arvidkahl: Tomorrow on the podcast and in the newsletter: why you should never start a SaaS business. A couple of really painful lesso…

RT @arvidkahl: Tomorrow on the podcast and in the newsletter: why you should never start a SaaS business. A couple of really painful lesso…

Building https://t.co/od97B0HVrk and https://t.co/666FnyVVE0 in Public. Raising all the boats with kindness. 🎙️ https://t.co/6w69DZmi8H · ✍️ https://t.co/lpnor5rsTW

avatar for Arvid Kahl
Arvid Kahl
Fri Oct 31 07:38:18
Josiah now at 561 followers, orbital injection successful I am always happy to launch another lowbie with alpha

Josiah now at 561 followers, orbital injection successful I am always happy to launch another lowbie with alpha

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

avatar for Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Fri Oct 31 07:37:16
  • Previous
  • 1
  • More pages
  • 1603
  • 1604
  • 1605
  • More pages
  • 2117
  • Next