BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.0.1//EN TZID:Europe/Paris X-WR-TIMEZONE:Europe/Paris BEGIN:VEVENT UID:253@isdm.umontpellier.fr DTSTART;TZID=Europe/Paris:20241021T140000 DTEND;TZID=Europe/Paris:20241021T140000 DTSTAMP:20260828T123738Z URL:https://isdm.umontpellier.fr/events/multi-task-partially-supervised-le arning-for-object-detection-and-semantic-segmentation/ SUMMARY:Multi-task partially supervised learning for object detection and s emantic segmentation DESCRIPTION:Room 02.124\, Building 5\, St Priest campus\n\nMachine Learning in Montpellier\, Theory &\; Practice\n\nComputer vision tasks are freq uently trained using distinctly annotated datasets although the tasks are often related to one another. Multi-task learning\, on the other hand\, is known to generally improve performance by leveraging the inter-task relat ionship but often at the high cost of maintaining all-task annotations for each training example. The multi-task partially supervised learning parad igm relaxes this requirement\, allowing each input to be annotated only fo r one of the target tasks. This setting hinders learning joint-task repres entations yet is potentially helpful for data scarcity scenarios as datase ts for individual tasks can be combined for training. In this presentation \, we study multi-task relationships for better data exploitation under th e partial supervision assumption for the two well-known computer vision ta sks\, namely object detection and semantic segmentation. Both of these tas ks are designed for scene understanding yet differ in data structure and i nformation level: object detection requires box coordinates for object ins tances while semantic segmentation requires pixel-wise regional categories . To that end\, we propose Box-for-Mask and Mask-for-Box strategies to ext ract information from one task's annotations to train the other. Ablation studies and experimental results on VOC and COCO datasets show favorable r esults for the proposed idea.\n\nMachine Learning in Montpellier\, Theory &\; Practice ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/ 2026/06/ml-mtp-gC78d5.png CATEGORIES:ML MTP LOCATION:Room 02.124 Building 5\, St Priest Campus\, Montpellier\, 34000\, France X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=St Priest Campus\, Montpell ier\, 34000\, France;X-APPLE-RADIUS=100;X-TITLE=Room 02.124 Building 5:geo :0,0 END:VEVENT BEGIN:VTIMEZONE TZID:Europe/Paris X-LIC-LOCATION:Europe/Paris BEGIN:DAYLIGHT DTSTART:20240331T030000 TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST END:DAYLIGHT END:VTIMEZONE END:VCALENDAR