Code and data
Data, code and models released by NAVER LABS Europe.
| Title | Description | Year | Code/data | Related papers/Blogs |
|---|---|---|---|---|
| CroCo | CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completion | 2022 | GitHub | NeurIPS 2022 |
| Expohedron | Efficient pareto-optimal fairness-utility amortizations in repeated rankings | 2022 | GitHub | WSDM 2022/ Download |
| Learning super-features for image retrieval | Code for running our FIRe model. | 2022 | GitHub | ICLR 2022 |
| Multilingual machine translation | What do compressed multilingual machine translation models forget? | 2022 | GitHub | arXiv.org |
| PoseBERT | A novel, plug and play model for human 3D shape estimation in videos. | 2022 | GitHub | Blog |
| SMaLL-100 | SMaLL-100: introducing shallow multilingual machine translation model for low-resource languages | 2022 | GitHub | EMNLP 2022 |
| Single-Step Adversarial Training | Make Some Noise: Reliable and Efficient Single-Step Adversarial Training . | 2022 | GitHub | Publication |
| Zero-shot task generalization | Multitask prompted training | 2022 | GitHub promptsource GitHub t-zero | ICLR 2022 |
| Controlled natural language generation | Debiasing large pretrained language models using distributional control. | 2021 | GitHub | ICLR 2021/ Blog |
| Large-scale localization | Large-scale localization datasets in crowded indoor spaces | 2021 | Open Dataset | CVPR 2021 |
| NMT | Code, model checkpoints, test sets and outputs for 4 multilingual NMT papers (about efficient inference, continual learning, unsupervised NMT and domain adaptation) | 2021 | Code | Blog/ EMNLP 2021/ EMNLP 2021/ WMT 2021/ WMT 2021 |
| SPLADE | A sparse bi-encoder BERT-based model achieves effective and efficient first-stage ranking | 2021 | GitHub | SIGIR 2021/ Blog GitHub Available Models Hugging Face |
| StacMR | Scene-Text Aware Cross-Modal Retrieval | 2021 | GitHub | Explore data, CTC / WACV ’21 |
| TLDR | Twin Learning for Dimensionality Reduction. TLDR is a method that is simple, easy to implement and train, and of broad applicability. | 2021 | GitHub | arXiv |
| Visual representation learning | Concept generalization in visual representation learning | 2021 | GitHub | ICCV 2021 / ICCV 2021 |
| COVID-19 NMT | Multi-lingual & multi-domain (specialisation for biomedical data) translation model. | 2020 | GitHub | Blog |
| DCMM | Differentiable Cross Modal Model (the code implementing the model introduced in Learning to Rank Images with Cross-Modal Graph Convolutions, ECIR'20). | 2020 | GitHub | ECIR 2020 / Blog |
| DOPE | Distillation of Part Experts for whole-body 3D pose estimation in the wild. | 2020 | GitHub | ECCV 2020/ Blog |
| FORCE | Progressive Skeletonization: Trimming more fat from a network at initialization. | 2020 | GitHub | arXiv/ Blog |
| Kapture | Kapture is a file format as well as a set of tools for manipulating datasets, and in particular Visual Localization and Structure from Motion data. | 2020 | GitHub | arXiv/ Blog |
| Kapture localization | A toolbox with various localization related algorithms (mapping, localization, benchmarking IR for VL). Relies strongly on the kapture format for data representation and manipulation. | 2020 | GitHub | 3DV 2020 |
| LCR-Net release V2.0 | Improved pose proposals integration for Multi-person 2D and 3D pose detection in natural images. | 2020 | GitHub | CVPR 2017/ TPAMI 2019 |
| LispE: Lisp Elémentaire | Ultra-minimal version of Lisp. Implementation of fully fledged Lisp interpreter with Data Structure, Pattern Programming and High level Functions with Lazy Evaluation à la Haskell. Comes with editor from TAMGU. | 2020 | GitHub | Code |
| MOCHI | Mixing of Contrastive Hard negatives (MOCHI), data mixing strategies that can be computed on-the-fly with minimal computational overhead, highly transferable visual representations. | 2020 | Download models | NeurIPS2020 |
| SMPLy | SMPLy benchmarking 3D human pose estimation in the wild. | 2020 | Dataset and code | Annotations |
| Virtual KITTI 2 | Updated photo-realistic synthetic video dataset designed to learn and evaluate computer vision models for several video understanding tasks: object detection and multi-object tracking, scene-level and instance-level semantic segmentation, optical flow, and depth estimation. | 2020 | Download dataset | arXiv/ Blog |
| Domain Shift Prediction | A method to predict the drop in accuracy of a trained model. | 2019 | GitHub | EMNLP2019/ Blog |
| MIMETICS | Understanding human action recognition. | 2019 | Dataset | arXiv/ Blog |
| Mallscape dataset | Dataset addresses all possible POI change scenarios to automatically update complex indoor maps. | 2019 | Download Mallscape A/ Download Mallscape B Evaluation code download here | CVPR 2019/ Blog |
| Motion-Augmented RGB Stream for Action Recognition | The test code and models are released under the MIT license. Both the code and models are on GitHub. | 2019 | Code | CVPR 2019/ Blog |
| R2D2 | Reliable and Repeatable Detector and Descriptor. | 2019 | GitHub (Code and data) | NeurIPs 2019/ Blog |
| Tamgu | A Functional, Imperative and Logical programming language for data annotation and augmentation. | 2019 | GitHub | Blog |
| Virtual Gallery Dataset | Synthetic dataset targeting challenges such as varying lighting conditions and different occlusion levels for tasks such as depth estimation, instance segmentation and visual localization. | 2019 | Dataset | CVPR 2019 |
| Aspect based sentiment analysis dataset | Manually annotated new ABSA dataset (different from Semeval2016) from Foursquare comments. | 2018 | Download dataset | WASSA 2018 |
| Regression applications | Theoretical and experimental findings to improve regression applications. | 2018 | GitHub | Blog/3DV 2021 |
| ReviewQA | A question-answering dataset based on hotel reviews. | 2018 | GitHub | Blog |
| Deep Image Retrieval | End-to-end learning of deep visual representations for image retrieval. | 2017 | Models and scripts of papers | ICCV 2019/ IJCV 2017/ Web |
| Procedural Human Action Videos dataset | A diverse, realistic, and physically plausible dataset of human action videos containing 39,982 videos, with more than 1,000 examples for each action of 35 categories. | 2017 | Dataset download | CVPR 2017 |