Tetiana Martyniuk
@t-martyniuk.bsky.social
a Ukrainian computer vision researcher escaping Twitter madness 😅
3D Computer Vision & Generative modelling
PhD student @ Inria Paris & @valeoai.bsky.social | Researcher @ Ukrainian Catholic University
https://t-martyniuk.github.io/
3D Computer Vision & Generative modelling
PhD student @ Inria Paris & @valeoai.bsky.social | Researcher @ Ukrainian Catholic University
https://t-martyniuk.github.io/
I just realized a few days ago they launched the bookmarks! Before I was doing some social media acrobatics to save the posts.
Do you see that many fake posts though? I feel like it’s indeed vital for current Twitter, but I don’t notice this problem as much here.
Do you see that many fake posts though? I feel like it’s indeed vital for current Twitter, but I don’t notice this problem as much here.
October 13, 2025 at 3:45 PM
I just realized a few days ago they launched the bookmarks! Before I was doing some social media acrobatics to save the posts.
Do you see that many fake posts though? I feel like it’s indeed vital for current Twitter, but I don’t notice this problem as much here.
Do you see that many fake posts though? I feel like it’s indeed vital for current Twitter, but I don’t notice this problem as much here.
Summary slides and Conclusion thoughts.
October 13, 2025 at 2:54 PM
Summary slides and Conclusion thoughts.
Yes, the flag-planting is a real race in the research, and I feel it’s even more incentivized with annual deadlines pressure that often leads to the submits that are extremely raw and unpolished yet quite novel. As I saw in one of the quotes of this post - ideas are cheap indeed.
October 11, 2025 at 8:56 PM
Yes, the flag-planting is a real race in the research, and I feel it’s even more incentivized with annual deadlines pressure that often leads to the submits that are extremely raw and unpolished yet quite novel. As I saw in one of the quotes of this post - ideas are cheap indeed.
Some conclusions and food for thought for the future research.
October 7, 2025 at 12:45 PM
Some conclusions and food for thought for the future research.
Some insights from soon-to-be Dr. Michele: Conclusions and Perspectives.
October 6, 2025 at 12:40 PM
Some insights from soon-to-be Dr. Michele: Conclusions and Perspectives.
I took my parents to see it, my mom even cried! She liked it a lot.
I liked it but with a small shortcoming that it was a bit like a fairytale: with his driving style he’d get 12 penalty points in the super license easily within those 9 races 🙈
I liked it but with a small shortcoming that it was a bit like a fairytale: with his driving style he’d get 12 penalty points in the super license easily within those 9 races 🙈
June 29, 2025 at 2:53 PM
I took my parents to see it, my mom even cried! She liked it a lot.
I liked it but with a small shortcoming that it was a bit like a fairytale: with his driving style he’d get 12 penalty points in the super license easily within those 9 races 🙈
I liked it but with a small shortcoming that it was a bit like a fairytale: with his driving style he’d get 12 penalty points in the super license easily within those 9 races 🙈
Did you like it?
June 29, 2025 at 7:52 AM
Did you like it?
LiDPM is a collaboration between Inria and @valeoai.bsky.social, couldn't have done it without the support of my amazing supervisors @alexandreboulch.bsky.social, @gillespuy.bsky.social, Renaud Marlet and Raoul de Charette!
June 25, 2025 at 8:07 PM
LiDPM is a collaboration between Inria and @valeoai.bsky.social, couldn't have done it without the support of my amazing supervisors @alexandreboulch.bsky.social, @gillespuy.bsky.social, Renaud Marlet and Raoul de Charette!
We show that "vanilla" diffusion can be applied to large-scale (here - lidar) non-normalized point clouds.
LiDPM
● does not require any additional regularization,
● inherently extends to unconditional scene generation,
● outperforms "local" diffusion for scene completion on SemanticKITTI.
LiDPM
● does not require any additional regularization,
● inherently extends to unconditional scene generation,
● outperforms "local" diffusion for scene completion on SemanticKITTI.
June 25, 2025 at 8:05 PM
We show that "vanilla" diffusion can be applied to large-scale (here - lidar) non-normalized point clouds.
LiDPM
● does not require any additional regularization,
● inherently extends to unconditional scene generation,
● outperforms "local" diffusion for scene completion on SemanticKITTI.
LiDPM
● does not require any additional regularization,
● inherently extends to unconditional scene generation,
● outperforms "local" diffusion for scene completion on SemanticKITTI.
It’s indeed a particular way to visualize the data :)
June 25, 2025 at 11:04 AM
It’s indeed a particular way to visualize the data :)
June 25, 2025 at 10:54 AM
June 25, 2025 at 10:52 AM
And a really great poster session afterwards! So many relevant questions, I was thrilled to share the details of our work, thanks for stopping by. #iv2025
June 23, 2025 at 4:39 PM
And a really great poster session afterwards! So many relevant questions, I was thrilled to share the details of our work, thanks for stopping by. #iv2025
Okay that was stressful 🥲
June 23, 2025 at 11:18 AM
Okay that was stressful 🥲
June 22, 2025 at 1:12 PM
Dr. Ignacio Alvarez overviewing Generative AI for synthetic scenario generations, automated HD mapping, behavior of traffic agents, and world modeling.
#IV2025
#IV2025
June 22, 2025 at 7:04 AM
Dr. Ignacio Alvarez overviewing Generative AI for synthetic scenario generations, automated HD mapping, behavior of traffic agents, and world modeling.
#IV2025
#IV2025
