Clément Rambour
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Clément Rambour has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Nicolas Thome
ISIR
7 shared publications - Elias Ramzi
CNAM
6 shared publications - Nicolas Audebert
CNAM
6 shared publications - Xavier Bitot
Coexya Paris France
3 shared publications - Andre Araujo
Google DeepMind
2 shared publications - Florence Tupin
Institut Polytechnique de Paris
2 shared publications
Latest Publications
- 1 October 2026
- Published by Springer Nature
- p. 538-552
Loc-Aware: Integrating Geographic Metadata into Contrastive Learning
- 1 October 2026
- Published by Springer Nature
- p. 538-552
Loc-Aware: Integrating Geographic Metadata into Contrastive Learning
- 3 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Structural tomographic approaches for urban area analysis using high resolution SAR tomography : TomoSAR
Approches tomographiques structurelles pour l'analyse du milieu urbain par tomographie SAR THR : TomoSAR: La tomographie SAR exploite plusieurs acquisitions d'une même zone acquises d'un point de vue légerement différent pour reconstruire la densité complexe de réflectivité au sol. Cette technique d'imagerie s'appuyant sur l'émission et la réception d'ondes électromagnétiques cohérentes, les données analysées sont complexes et l'information spatiale manquante (selon la verticale) est codée dans la phase. De nombreuse méthodes ont pu être proposées pour retrouver cette information. L'utilisation des redondances naturelles à certains milieux n'est toutefois généralement pas exploitée pour améliorer l'estimation tomographique. Cette thèse propose d'utiliser l'information structurelle propre aux structures urbaines pour régulariser les densités de réflecteurs obtenues par cette technique.
- 3 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Structural tomographic approaches for urban area analysis using high resolution SAR tomography : TomoSAR
Approches tomographiques structurelles pour l'analyse du milieu urbain par tomographie SAR THR : TomoSAR: La tomographie SAR exploite plusieurs acquisitions d'une même zone acquises d'un point de vue légerement différent pour reconstruire la densité complexe de réflectivité au sol. Cette technique d'imagerie s'appuyant sur l'émission et la réception d'ondes électromagnétiques cohérentes, les données analysées sont complexes et l'information spatiale manquante (selon la verticale) est codée dans la phase. De nombreuse méthodes ont pu être proposées pour retrouver cette information. L'utilisation des redondances naturelles à certains milieux n'est toutefois généralement pas exploitée pour améliorer l'estimation tomographique. Cette thèse propose d'utiliser l'information structurelle propre aux structures urbaines pour régulariser les densités de réflecteurs obtenues par cette technique.
- 20 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (6) , 4317-4329
Optimization of Rank Losses for Image Retrieval
In image retrieval, standard evaluation metrics rely on score ranking, e.g. average precision (AP), recall at k (R@k), normalized discounted cumulative gain (NDCG). In this work, we introduce a general framework for robust and decomposable rank losses optimization. It addresses two major challenges for end-to-end training of deep neural networks with rank losses: non-differentiability and non-decomposability. Firstly, we propose a general surrogate for ranking operator, SupRank, that is amenable to stochastic gradient descent. It provides an upperbound for rank losses and ensures robust training. Secondly, we use a simple yet effective loss function to reduce the decomposability gap between the averaged batch approximation of ranking losses and their values on the whole training set. We apply our framework to two standard metrics for image retrieval: AP and R@k. Additionally, we apply our framework to hierarchical image retrieval. We introduce an extension of AP, the hierarchical average precision ${\mathcal {H}}$, and optimize it as well as the NDCG. Finally, we create the first hierarchical landmarks retrieval dataset. We use a semi-automatic pipeline to create hierarchical labels, extending the large scale Google Landmarks v2 dataset. The hierarchical dataset is publicly available at github.com/cvdfoundation/google-landmark. Code is available at github.com/elias-ramzi/SupRank.
- 20 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (6) , 4317-4329
Optimization of Rank Losses for Image Retrieval
In image retrieval, standard evaluation metrics rely on score ranking, e.g. average precision (AP), recall at k (R@k), normalized discounted cumulative gain (NDCG). In this work, we introduce a general framework for robust and decomposable rank losses optimization. It addresses two major challenges for end-to-end training of deep neural networks with rank losses: non-differentiability and non-decomposability. Firstly, we propose a general surrogate for ranking operator, SupRank, that is amenable to stochastic gradient descent. It provides an upperbound for rank losses and ensures robust training. Secondly, we use a simple yet effective loss function to reduce the decomposability gap between the averaged batch approximation of ranking losses and their values on the whole training set. We apply our framework to two standard metrics for image retrieval: AP and R@k. Additionally, we apply our framework to hierarchical image retrieval. We introduce an extension of AP, the hierarchical average precision ${\mathcal {H}}$, and optimize it as well as the NDCG. Finally, we create the first hierarchical landmarks retrieval dataset. We use a semi-automatic pipeline to create hierarchical labels, extending the large scale Google Landmarks v2 dataset. The hierarchical dataset is publicly available at github.com/cvdfoundation/google-landmark. Code is available at github.com/elias-ramzi/SupRank.
- 24 November 2024
- Published by Springer Nature
- p. 264-282
GalLoP: Learning Global and Local Prompts for Vision-Language Models
- 24 November 2024
- Published by Springer Nature
- p. 264-282
GalLoP: Learning Global and Local Prompts for Vision-Language Models
- Open Access
VidEdit: Zero-Shot and Spatially Aware Text-Driven Video Editing
Recently, diffusion-based generative models have achieved remarkable success for image generation and edition. However, existing diffusion-based video editing approaches lack the ability to offer precise control over generated content that maintains temporal consistency in long-term videos. On the other hand, atlas-based methods provide strong temporal consistency but are costly to edit a video and lack spatial control. In this work, we introduce VidEdit, a novel method for zero-shot text-based video editing that guarantees robust temporal and spatial consistency. In particular, we combine an atlas-based video representation with a pre-trained text-to-image diffusion model to provide a training-free and efficient video editing method, which by design fulfills temporal smoothness. To grant precise user control over generated content, we utilize conditional information extracted from off-the-shelf panoptic segmenters and edge detectors which guides the diffusion sampling process. This method ensures a fine spatial control on targeted regions while strictly preserving the structure of the original video. Our quantitative and qualitative experiments show that VidEdit outperforms state-of-the-art methods on DAVIS dataset, regarding semantic faithfulness, image preservation, and temporal consistency metrics. With this framework, processing a single video only takes approximately one minute, and it can generate multiple compatible edits based on a unique text prompt. Project web-page at https://videdit.github.io
Open AccessVidEdit: Zero-Shot and Spatially Aware Text-Driven Video Editing
Recently, diffusion-based generative models have achieved remarkable success for image generation and edition. However, existing diffusion-based video editing approaches lack the ability to offer precise control over generated content that maintains temporal consistency in long-term videos. On the other hand, atlas-based methods provide strong temporal consistency but are costly to edit a video and lack spatial control. In this work, we introduce VidEdit, a novel method for zero-shot text-based video editing that guarantees robust temporal and spatial consistency. In particular, we combine an atlas-based video representation with a pre-trained text-to-image diffusion model to provide a training-free and efficient video editing method, which by design fulfills temporal smoothness. To grant precise user control over generated content, we utilize conditional information extracted from off-the-shelf panoptic segmenters and edge detectors which guides the diffusion sampling process. This method ensures a fine spatial control on targeted regions while strictly preserving the structure of the original video. Our quantitative and qualitative experiments show that VidEdit outperforms state-of-the-art methods on DAVIS dataset, regarding semantic faithfulness, image preservation, and temporal consistency metrics. With this framework, processing a single video only takes approximately one minute, and it can generate multiple compatible edits based on a unique text prompt. Project web-page at https://videdit.github.io
- Open Access
Energy Correction Model in the Feature Space for Out-of-Distribution Detection
In this work, we study the out-of-distribution (OOD) detection problem through the use of the feature space of a pre-trained deep classifier. We show that learning the density of in-distribution (ID) features with an energy-based models (EBM) leads to competitive detection results. However, we found that the non-mixing of MCMC sampling during the EBM's training undermines its detection performance. To overcome this an energy-based correction of a mixture of class-conditional Gaussian distributions. We obtains favorable results when compared to a strong baseline like the KNN detector on the CIFAR-10/CIFAR-100 OOD detection benchmarks.
Open AccessEnergy Correction Model in the Feature Space for Out-of-Distribution Detection
In this work, we study the out-of-distribution (OOD) detection problem through the use of the feature space of a pre-trained deep classifier. We show that learning the density of in-distribution (ID) features with an energy-based models (EBM) leads to competitive detection results. However, we found that the non-mixing of MCMC sampling during the EBM's training undermines its detection performance. To overcome this an energy-based correction of a mixture of class-conditional Gaussian distributions. We obtains favorable results when compared to a strong baseline like the KNN detector on the CIFAR-10/CIFAR-100 OOD detection benchmarks.
- Open Access
- 1 February 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 241
MERLIN-Seg: Self-supervised despeckling for label-efficient semantic segmentation
- 1 February 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 241
Open AccessMERLIN-Seg: Self-supervised despeckling for label-efficient semantic segmentation
- Open Access
Optimization of Rank Losses for Image Retrieval
In image retrieval, standard evaluation metrics rely on score ranking, \eg average precision (AP), recall at k (R@k), normalized discounted cumulative gain (NDCG). In this work we introduce a general framework for robust and decomposable rank losses optimization. It addresses two major challenges for end-to-end training of deep neural networks with rank losses: non-differentiability and non-decomposability. Firstly we propose a general surrogate for ranking operator, SupRank, that is amenable to stochastic gradient descent. It provides an upperbound for rank losses and ensures robust training. Secondly, we use a simple yet effective loss function to reduce the decomposability gap between the averaged batch approximation of ranking losses and their values on the whole training set. We apply our framework to two standard metrics for image retrieval: AP and R@k. Additionally we apply our framework to hierarchical image retrieval. We introduce an extension of AP, the hierarchical average precision $\mathcal{H}$-AP, and optimize it as well as the NDCG. Finally we create the first hierarchical landmarks retrieval dataset. We use a semi-automatic pipeline to create hierarchical labels, extending the large scale Google Landmarks v2 dataset. The hierarchical dataset is publicly available at https://github.com/cvdfoundation/google-landmark. Code will be released at https://github.com/elias-ramzi/SupRank.
Open AccessOptimization of Rank Losses for Image Retrieval
In image retrieval, standard evaluation metrics rely on score ranking, \eg average precision (AP), recall at k (R@k), normalized discounted cumulative gain (NDCG). In this work we introduce a general framework for robust and decomposable rank losses optimization. It addresses two major challenges for end-to-end training of deep neural networks with rank losses: non-differentiability and non-decomposability. Firstly we propose a general surrogate for ranking operator, SupRank, that is amenable to stochastic gradient descent. It provides an upperbound for rank losses and ensures robust training. Secondly, we use a simple yet effective loss function to reduce the decomposability gap between the averaged batch approximation of ranking losses and their values on the whole training set. We apply our framework to two standard metrics for image retrieval: AP and R@k. Additionally we apply our framework to hierarchical image retrieval. We introduce an extension of AP, the hierarchical average precision $\mathcal{H}$-AP, and optimize it as well as the NDCG. Finally we create the first hierarchical landmarks retrieval dataset. We use a semi-automatic pipeline to create hierarchical labels, extending the large scale Google Landmarks v2 dataset. The hierarchical dataset is publicly available at https://github.com/cvdfoundation/google-landmark. Code will be released at https://github.com/elias-ramzi/SupRank.
- 16 December 2022
- Published by Springer Nature
- p. 121-130
Memory Transformers for Full Context and High-Resolution 3D Medical Segmentation
- 16 December 2022
- Published by Springer Nature
- p. 121-130
Memory Transformers for Full Context and High-Resolution 3D Medical Segmentation
- 23 October 2022
- Published by Springer Nature
- p. 250-266
Hierarchical Average Precision Training for Pertinent Image Retrieval
- 23 October 2022
- Published by Springer Nature
- p. 250-266
Hierarchical Average Precision Training for Pertinent Image Retrieval