--- title: "Alaaeldin El-Nouby — author profile and papers" description: "Alaaeldin El-Nouby: 2 papers on Elamur Research (2023) — affiliations, topics, co-authors and who cites them in our library." canonical: "https://research.elamur.ai/free/en/authors/alaaeldin-el-nouby/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/alaaeldin-el-nouby/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Alaaeldin El-Nouby # Alaaeldin El-Nouby Alaaeldin El-Nouby works on learning general-purpose representations for computer vision. The two papers here, both from 2023, ask how far large-scale pretraining can go without hand-made labels. One question is whether image features can be reused across many tasks. Another is whether several kinds of sensor signal can share a single embedding space. The methods are Transformer encoders, contrastive objectives and very large image collections. The more cited of the two is [DINOv2: Learning Robust Visual Features without Supervision](https://md.research.elamur.ai/free/en/papers/oquab-2023-dinov2-learning-robust-visual-features/) (2023). It tests whether self-supervised pretraining by itself can yield frozen image features that transfer to varied tasks without finetuning and hold up against text-supervised encoders such as CLIP. To that end, the paper assembles a 142-million-image dataset, LVD-142M, from the web. The pipeline deduplicates images and keeps those that resemble images in curated sets, and it uses neither captions nor metadata. The second paper, [ImageBind: One Embedding Space To Bind Them All](https://md.research.elamur.ai/free/en/papers/girdhar-2023-imagebind-one-embedding-space-bind/) (2023), deals with the lack of any dataset that captures every modality at once. It uses images as a common anchor and links six modalities: images and video, text, audio, depth, thermal and IMU readings. Each non-image encoder is trained with a contrastive loss that draws it toward the embedding of its paired image. Both papers are also the most recent ones here. Together they point from single-modality visual features toward joint spaces that span many sensory inputs. Co-authors on these papers include [Armand Joulin](https://md.research.elamur.ai/free/en/authors/armand-joulin/), [Ishan Misra](https://md.research.elamur.ai/free/en/authors/ishan-misra/), [Nicolas Ballas](https://md.research.elamur.ai/free/en/authors/nicolas-ballas/) and [Francisco Massa](https://md.research.elamur.ai/free/en/authors/francisco-massa/), along with Daniel Haziza and Zhuang Liu. Alaaeldin El-Nouby has 2 papers in our library, in 2023; often with [Armand Joulin](https://md.research.elamur.ai/free/en/authors/armand-joulin/), [Daniel Haziza](https://md.research.elamur.ai/free/en/authors/daniel-haziza/), [Francisco Massa](https://md.research.elamur.ai/free/en/authors/francisco-massa/). 2 papers in our library · 2023 No OpenAlex profile matched yet. [2 works in our library](https://md.research.elamur.ai/free/en/authors/alaaeldin-el-nouby/#papers) · [467 citations in our library](https://md.research.elamur.ai/free/en/authors/alaaeldin-el-nouby/cited-by/) · [⭐ Paper Star rank #1036](https://md.research.elamur.ai/free/en/paper-stars/page/11/#star-1036) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Alaaeldin+El-Nouby) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Alaaeldin+El-Nouby) · [dblp](https://dblp.org/search?q=Alaaeldin+El-Nouby) · [ACL Anthology](https://aclanthology.org/search/?q=Alaaeldin+El-Nouby) ## Works by year ### 2020s [**2023** 2 papers](https://md.research.elamur.ai/free/en/authors/alaaeldin-el-nouby/year/2023/) ## Papers The most cited first (OpenAlex and our library). 1. **[DINOv2: Learning Robust Visual Features without Supervision](https://md.research.elamur.ai/free/en/papers/oquab-2023-dinov2-learning-robust-visual-features/)** *(2023 · 1,425 citations)* Self-supervised vision models have mostly been trained either on the small ImageNet-1k set or on large, noisy web collections, and both choices limit how general their features are. 2. **[ImageBind: One Embedding Space To Bind Them All](https://md.research.elamur.ai/free/en/papers/girdhar-2023-imagebind-one-embedding-space-bind/)** *(2023 · 943 citations)* ImageBind addresses a practical obstacle to building one representation space for many sensory modalities: no large dataset records all modalities together. ## Co-authors here 6 co-authors with a page, the most papers together first. - [Armand Joulin](https://md.research.elamur.ai/free/en/authors/armand-joulin/) - [Daniel Haziza](https://md.research.elamur.ai/free/en/authors/daniel-haziza/) - [Francisco Massa](https://md.research.elamur.ai/free/en/authors/francisco-massa/) - [Ishan Misra](https://md.research.elamur.ai/free/en/authors/ishan-misra/) - [Nicolas Ballas](https://md.research.elamur.ai/free/en/authors/nicolas-ballas/) - [Zhuang Liu](https://md.research.elamur.ai/free/en/authors/zhuang-liu/) ## Cited by Papers of our library cite their works 467 times (of the 140,884 papers whose bibliographies we read); [14 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/alaaeldin-el-nouby/cited-by/).