Bersama dengan ini, kami umumkan kegiatan Seminar Reboan yang dilaksanakan pada:
| Tanggal | : | Rabu, 20 November 2019 |
| Waktu | : | 13:00 – 14:00 WIB |
| Tempat | : | Ruang 1213, Gedung A lantai 2, Fasilkom UI Depok |
| Topik | : | Super-resolution for Learning Visual Representations: Application to Tiny Object Detection |
| Pembicara | : | Dr. Eng. Muhammad Haris (Postdoctoral fellow di Intelligent Information Media Laboratory, Toyota Technological Institute) |
Abstrak:
Super-Resolution (SR) is an ill-posed inverse problem where the aim is to recover a high-resolution (HR) image from a low-resolution (LR) image. We consider how image super-resolution (SR) can contribute to an object detection task in low-resolution images. Intuitively, SR gives a positive impact on the object detection task. While several previous works demonstrated that this intuition is correct, SR and detector are optimized independently in these works. This paper proposes a novel framework to train a deep neural network where the SR sub-network explicitly incorporates a detection loss in its training objective, via a tradeoff with a traditional detection loss. This end-to-end training procedure allows us to train SR preprocessing for any differentiable detector. We demonstrate extensive experiments that show our task-driven SR consistently and significantly improves the accuracy of an object detector on low-resolution images for a variety of conditions and scaling factors.
Apabila ada diantara bapak/ibu yang berminat hadir, kami persilahkan mendaftarkan diri (RSVP) pada : https://bit.ly/rsvpfasilkomui-1
Jika ada pertanyaan silahkan disamapikan ke email: humasfasilkom[at]cs.ui.ac.id