kjyong https://kjyong.inblog.io On-device AI Computer Vision Engineerhttps://www.linkedin.com/in/jayong-kim-60ba80121/ Sun, 02 Jun 2024 03:01:27 GMT inblog Sun, 29 Mar 2026 14:27:14 GMT https://image.inblog.dev?url=https%3A%2F%2Finblog.ai%2Fapi%2Fog-logo%3FlogoUrl%3Dhttps%253A%252F%252Finblog.ai%252Finblog_logo.png%26title%3Dkjyong%26description%3DOn-device%2520AI%2520%2508Computer%2520Vision%2520Engineer%250Ahttps%253A%252F%252Fwww.linkedin.com%252Fin%252Fjayong-kim-60ba80121%252F&w=1920&q=85 og image https://kjyong.inblog.io MagicVideo: Efficient Video Generation With Latent Diffusion Models https://kjyong.inblog.io/magicvideo-efficient-video-generation-with-latent-diffusion-models-149413 Sun, 29 Mar 2026 14:27:33 GMT https://kjyong.inblog.io/magicvideo-efficient-video-generation-with-latent-diffusion-models-149413 바세린용자 (yongja) Video Diffusion Models (VDM) https://kjyong.inblog.io/video-diffusion-models-vdm-128700 Sun, 15 Mar 2026 15:00:44 GMT https://kjyong.inblog.io/video-diffusion-models-vdm-128700 바세린용자 (yongja) InstantID: Zero-shot Identity Preserving Generation in Seconds https://kjyong.inblog.io/instantid-zeroshot-identity-preserving-generation-in-seconds-99051 Sun, 01 Feb 2026 12:44:33 GMT https://kjyong.inblog.io/instantid-zeroshot-identity-preserving-generation-in-seconds-99051 바세린용자 (yongja) SnapFusion: Text-to-Image Diffusion Model on MobileDevices within Two Seconds https://kjyong.inblog.io/snapfusion-texttoimage-diffusion-model-on-mobile-devices-within-two-seconds-97109 Sat, 24 Jan 2026 14:05:26 GMT https://kjyong.inblog.io/snapfusion-texttoimage-diffusion-model-on-mobile-devices-within-two-seconds-97109 바세린용자 (yongja) LoRA: Low-Rank Adaptation of Large Language Models https://kjyong.inblog.io/lora-lowrank-adaptation-of-large-language-models-94786 Sat, 17 Jan 2026 14:32:18 GMT https://kjyong.inblog.io/lora-lowrank-adaptation-of-large-language-models-94786 바세린용자 (yongja) Textual Inversion: An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion https://kjyong.inblog.io/textual-inversion-an-image-is-worth-one-word-personalizing-texttoimage-generation-using-textual-inversion-93184 Sun, 11 Jan 2026 13:15:36 GMT https://kjyong.inblog.io/textual-inversion-an-image-is-worth-one-word-personalizing-texttoimage-generation-using-textual-inversion-93184 바세린용자 (yongja) DreamBooth: Fine Tuning Text-to-Image Diffusion Modelsfor Subject-Driven Generation https://kjyong.inblog.io/dreambooth-fine-tuning-texttoimage-diffusion-models-for-subjectdriven-generation-78630 Sun, 04 Jan 2026 10:26:07 GMT https://kjyong.inblog.io/dreambooth-fine-tuning-texttoimage-diffusion-models-for-subjectdriven-generation-78630 바세린용자 (yongja) LDM: High-Resolution Image Synthesis with Latent Diffusion Models https://kjyong.inblog.io/ldm-highresolution-image-synthesis-with-latent-diffusion-models-87232 Sun, 21 Dec 2025 05:24:08 GMT https://kjyong.inblog.io/ldm-highresolution-image-synthesis-with-latent-diffusion-models-87232 바세린용자 (yongja) VQ-VAE: Neural Discrete Representation Learning https://kjyong.inblog.io/vqvae-85312 Sun, 14 Dec 2025 12:21:38 GMT https://kjyong.inblog.io/vqvae-85312 컴퓨터비전 바세린용자 (yongja) MAE와 MSE는 부호를 없애는데 어떻게 regression이 학습되는 걸까? https://kjyong.inblog.io/mae와-mse는-부호를-없애는데-어떻게-regression이-학습되는-걸까-23162 regression loss의 미분값에 대한 이해 Sat, 29 Jun 2024 09:27:00 GMT https://kjyong.inblog.io/mae와-mse는-부호를-없애는데-어떻게-regression이-학습되는-걸까-23162 딥러닝 바세린용자 (yongja) 왜 Re-Identification에서 GeM을 쓸까? https://kjyong.inblog.io/왜-reidentification에서-gem을-쓸까-23161 GeM Pooling VS Max Pooling vs Average Pooling Sat, 29 Jun 2024 09:11:00 GMT https://kjyong.inblog.io/왜-reidentification에서-gem을-쓸까-23161 컴퓨터비전딥러닝 바세린용자 (yongja) [Knowledge Distillation] Learning from a Teacher using Unlabeled Data https://kjyong.inblog.io/21254 Unlabelling 데이터를 활용하여 Teacher보다 똑똑한 Student 만들기 Sun, 02 Jun 2024 03:09:00 GMT https://kjyong.inblog.io/21254 컴퓨터비전경량화딥러닝 바세린용자 (yongja)