Cédric Picron

PhD candidate in Computer Vision

KU Leuven

Biography

Cédric Picron is a PhD candidate at the KU Leuven ESAT-PSI lab under supervision of Prof. Tinne Tuytelaars. His research interests include designing efficient and high-performing network architectures for various computer vision tasks, such as object detection and instance segmentation.

Interests

Architecture Design
Computer Vision
Object Detection
Segmentation

Education

PhD in Computer Vision, 2023 (ongoing)
KU Leuven
MSc in Mathemetical Engineering, 2019
KU Leuven
BSc in Electrical Engineering, 2017
Vrije Universiteit Brussel

Featured Research

Cédric Picron, Tinne Tuytelaars

July, 2023 In ArXiv

EffSeg: Efficient Fine-Grained Instance Segmentation using Structure-Preserving Sparsity

A new instance segmentation head performing fine-grained instance segmentation in an efficient way by using our Structure-Preserving Sparsity (SPS) method based on separately storing the active features, the passive features and a dense 2D index map containing the feature indices.

Recent Research

Cédric Picron, Punarjay Chakravarty, Tinne Tuytelaars (2022). FQDet: Fast-converging Query-based Detector. In NeurIPS VTTA workshop 2022.

PDF Cite Code Poster

Cédric Picron, Tinne Tuytelaars (2022). Trident Pyramid Networks for Object Detection. In BMVC 2022.

PDF Cite Code Poster

Georgios Kouros, Shubham Shrivastava, Cédric Picron, Sushruth Nagesh, Punarjay Chakravarty, Tinne Tuytelaars (2022). Category-level pose retrieval with contrastive features learnt with occlusion augmentation. In BMVC 2022.

PDF Cite Code Poster

Cédric Picron, Punarjay Chakravarty, Tom Roussel, Tinne Tuytelaars (2021). What My Motion tells me about Your Pose: A Self-Supervised Monocular 3D Vehicle Detector. In ICRA 2021.

PDF Cite Code