Emre Akbas

@PINguAR Congratulations! Well deserved.

Thrilled to receive an NSF CAREER Award! Excited to push the frontiers of video generation toward controllable, cinematic tools for creating dynamic worlds. Grateful to my students, mentors, collaborators, and NSF. nsf.gov/awardsearch/sh…

10 yıldır Bilim Akademisi öncülüğünde sürdürülen Yapay Zekâ Yaz Okulu bu yıl 3-6 Temmuz arasında Piri Reis Üniversitesi Deniz Kampüsü’nde yapılacak! Bu yılın teması “Yapay Zekâda Temellerden Otonom Sistemlere ve Güvenilir Yapay Zekâya”. Ayrıntılar için: bilimakademisi.org/yaz-okulu-bayz…

1/ Explaining CLIP Zero-shot Predictions Through Concepts Onat Ozdemir, Anders Christensen, Stephan Alaniz, @zeynepakata, @eakbas2 [Paper]: arxiv.org/abs/2603.28211 [Project]: oonat.github.io/ezpc/ [Code]: github.com/oonat/ezpc

2/ Rethinking Concept Bottleneck Models: From Pitfalls to Solutions @merve_tapli , @QBouniot, @WolfStammer, @zeynepakata, @eakbas2 [Paper]: arxiv.org/pdf/2603.05629

@PaulGavrikov Looks like it wants to listen to music while vibe coding :-D

Uhm, why does Codex need access to my media library to fix my inference code???

Postdoc fellowship opportunities at @METU_ODTU: enrichtogether.metu.edu.tr for 24 months with good compensation.

Yapay Zekâ Yaz Okulu 3–6 Temmuz arasında Piri Reis Üniversitesinde yapılacak. Bu yılın teması “Yapay Zekâda Temellerden Otonom Sistemlere ve Güvenilir Yapay Zekâya”. Son kayıt tarihi 18 Mayıs. Dersler ve kayıt ile ilgili ayrıntılar bağlantıda. bilimakademisi.org/yaz-okulu-bayz…

@jhugestar Looks very good! You may find this CVPR22 tutorial useful: youtube.com/watch?v=UuB7hL… sites.google.com/view/performan…

I’m currently teaching a Computer Vision course, and I’ve found that explaining object detection evaluation, especially mAP, is not always straightforward. I tried using AI to build a small interactive tool that makes the mAP metric much more intuitive. It was a fun experience and also shows how useful AI can be for creating teaching materials! Sharing it here in case it’s helpful: jhugestar.github.io/snu-vision-lec…

Çiçeksepeti'nden bir düğün çelengi sipariş etmiştik. Görsel 1: web sitesinde bizim seçtiğimiz tasarım. Görsel 2: nikah günü çiçeksepeti'nin bize gönderdiği foto. Yazıları bulanıklaştırdım. Görsel 3: teslim edilen çelenk. Bir açıklamanız var mı @ciceksepeti ?

Links to the papers, code, and short summaries will be shared soon.

3. Explaining CLIP Zero-Shot Predictions Through Concepts. ✍ Onat Özdemir, Anders Christensen, Stephan Alaniz, Zeynep Akata, Emre Akbas. (Collaboration between @HelmholtzMunich and @METU_ODTU )

We will present the following papers at hashtag #CVPR 2026 in Denver, CO: 1. MatchED: Crisp Edge Detection Using End-to-End, Matching-Based Supervision. ✍ Bedrettin Çetinkaya, Sinan Kalkan, Emre Akbas.

@jbhuang0604 🤔 I don't remember this painting at all! 😁

Visited my office as a student 10 years ago at Beckman Institute! I always feel that this painting was dark and depressing (particularly so when I could not fix the bugs in my code). It’s still there!

@ftm_guney @metin_sitti Tebrikler!

Kadir Has Gelecek Vaadeden Bilim İnsanı Ödülü almışım 🤩 ne güzel haber, ne kadar anlamlı, bir o kadar da sorumluluk yükleyen bir ödül! değerli rektörümüz @metin_sitti Hoca'mızın kıymetli desteğiyle, ilk kez kendim başvurmadan ve birkaç kez reddedilmeden bir ödül aldım 😊

🔗 Course webpage: user.ceng.metu.edu.tr/~emre/Fall2025… The material may be useful for graduate students and researchers interested in deep learning fundamentals.

I am sharing the lecture recordings of CENG501 Deep Learning. 🎥 Lecture playlist: youtube.com/playlist?list=… (They are slides + audio only -- no instructor acting 🙂). The course page includes slides, Colab notebooks, and recommended readings:

Our genetic kinship estimation tool using deep learning is just out: onlinelibrary.wiley.com/doi/10.1111/17… Our models trained on simulated data work effectively on real data from diverse backgrounds and often better than available tools. Congratulations @mmerfff and team!

Our genetic kinship estimation tool “DeepKin” is now available! Our neural network models trained on simulated data work effectively on real ancient data from diverse backgrounds and often outperform available tools. onlinelibrary.wiley.com/doi/10.1111/17…