Пятый рождественский коллоквиум по компьютерному зрению / запись трансляции

4:57 — Large Scale Learning of General Visual Representations for Transfer - Александр Колесников 48:52 — AdaptIS: Adaptive Instance Selection Network - Константин Софиюк 1:22:43 — Unsupervised Neural Quantization for Compressed-Domain Similarity Search - Станислав Морозов 2:04:25 — 3D Deep Learning: Methods and Applications - Максим Татарченко 2:43:15 — Probabilistic Modeling for 3D Shape Generation and Inference-Based Reconstruction From a Single Image - Роман Клоков 3:13:58 — Perceptual Deep Depth Super-Resolution - Олег Войнов 4:01:24 — Random Paths as a Latent Space for GAN Interpretability - Андрей Войнов 4:28:51 — You Only Train Once: Loss-Conditional Training of Deep Networks - Алексей Досовицкий 4:53:23 — The Point Where Reality Meets Fantasy: Mixed Adversarial Generators for Image Splice Detection - Владимир Князь