<?xml version="1.0" encoding="UTF-8" ?>
<rss
    version="2.0"
    xmlns:atom="http://www.w3.org/2005/Atom"
    xmlns:content="http://purl.org/rss/1.0/modules/content/"
    xmlns:webfeeds="http://webfeeds.org/rss/1.0"
    xmlns:media="http://search.yahoo.com/mrss/"
    >
    <channel>
        <title>Computer-vision Tag - Viblo</title>
        <link>https://viblo.asia/rss</link>
        <description><![CDATA[Free service for technical knowledge sharing]]></description>
        <atom:link href="https://viblo.asia/rss/tags/computer-vision.rss" rel="self"></atom:link>
                <copyright>Sun* Inc.</copyright>
                                                <webfeeds:logo>https://viblo.asia/logo_full.svg</webfeeds:logo>
        <image>
            <url>https://viblo.asia/logo_full.svg</url>
            <title>Computer-vision Tag - Viblo</title>
            <link>https://viblo.asia/rss</link>
        </image>
                                <language>vi-vn</language>
        <lastBuildDate>2026-04-18T08:37:26+07:00</lastBuildDate>
                <item>
            <title><![CDATA[SADTALER - XÂY DỰNG BỨC ẢNH BIẾT NÓI]]></title>
                        <link>https://viblo.asia/p/sadtaler-xay-dung-buc-anh-biet-noi-bNVQGGoAJvR</link>
            <guid isPermaLink="true">https://viblo.asia/p/sadtaler-xay-dung-buc-anh-biet-noi-bNVQGGoAJvR</guid>
            <description><![CDATA[Lời mở đầu
Chắc hẳn bạn đã từng ao ước có thể "thổi hồn" vào những bức ảnh tĩnh, biến chúng trở nên sống động và biết nói? Hay bạn đang tìm kiếm một c...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trần Trung Kiên</dc:creator>
            <pubDate>2025-10-04 22:39:39</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[GStreamer cho người mới bắt đầu - Phần 2: Cài đặt, chạy, debug pipeline GStreamer với Python]]></title>
                        <link>https://viblo.asia/p/gstreamer-cho-nguoi-moi-bat-dau-phan-2-cai-dat-chay-debug-pipeline-gstreamer-voi-python-yZJZlKlNVjm</link>
            <guid isPermaLink="true">https://viblo.asia/p/gstreamer-cho-nguoi-moi-bat-dau-phan-2-cai-dat-chay-debug-pipeline-gstreamer-voi-python-yZJZlKlNVjm</guid>
            <description><![CDATA[Trong phần 1, mình đã trình bày lý thuyết về các thành phần cơ bản trong một GStreamer pipeline rồi, giờ thì bắt tay vào phần thực hành nào 

À vì các...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Lê Minh Tú</dc:creator>
            <pubDate>2025-07-03 12:38:43</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[GStreamer cho người mới bắt đầu - Phần 1: Giới thiệu thuật ngữ.]]></title>
                        <link>https://viblo.asia/p/gstreamer-cho-nguoi-moi-bat-dau-phan-1-gioi-thieu-thuat-ngu-Ny0VG2jpLPA</link>
            <guid isPermaLink="true">https://viblo.asia/p/gstreamer-cho-nguoi-moi-bat-dau-phan-1-gioi-thieu-thuat-ngu-Ny0VG2jpLPA</guid>
            <description><![CDATA[Khi chạy các bài toán Computer Vision cần phải xử lý các stream camera, mình hay đọc stream và lưu video bằng OpenCV. Vì nó đơn giản, dễ dùng, và code...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Lê Minh Tú</dc:creator>
            <pubDate>2025-06-18 20:03:52</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Image stitching]]></title>
                        <link>https://viblo.asia/p/image-stitching-EoW4oQXkLml</link>
            <guid isPermaLink="true">https://viblo.asia/p/image-stitching-EoW4oQXkLml</guid>
            <description><![CDATA[ Giới thiệu
Image Stitching hay còn được biết đến là ghép ảnh là một kỹ thuật trong thị giác máy tính nhằm kết hợp ảnh của cùng một khung cảnh nhưng đ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Duong Nguyen</dc:creator>
            <pubDate>2025-04-28 12:57:39</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Vintern-1B-v3.5] Mô hình kết hợp OCR và xử lý văn bản nhỏ nhẹ dành cho tiếng Việt]]></title>
                        <link>https://viblo.asia/p/vintern-1b-v35-mo-hinh-ket-hop-ocr-va-xu-ly-van-ban-nho-nhe-danh-cho-tieng-viet-AZoJj8WOVY7</link>
            <guid isPermaLink="true">https://viblo.asia/p/vintern-1b-v35-mo-hinh-ket-hop-ocr-va-xu-ly-van-ban-nho-nhe-danh-cho-tieng-viet-AZoJj8WOVY7</guid>
            <description><![CDATA[1. Giới thiệu

Vintern-1B-v3.5 là phiên bản mô hình ngôn ngữ lớn đa phương thức (MLLM) trong dòng mô hình Vintern được phát triển bởi nhóm nghiên cứu ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyễn Văn Mạnh</dc:creator>
            <pubDate>2025-03-31 13:50:24</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Papers Notes] NaViT: Thuật toán Transformer với Tầm nhìn Đa phân giải - Cải thiện khả năng nhìn của Transformer trong xử lý hình ảnh]]></title>
                        <link>https://viblo.asia/p/papers-notes-navit-thuat-toan-transformer-voi-tam-nhin-da-phan-giai-cai-thien-kha-nang-nhin-cua-transformer-trong-xu-ly-hinh-anh-GAWVpZ0XJ05</link>
            <guid isPermaLink="true">https://viblo.asia/p/papers-notes-navit-thuat-toan-transformer-voi-tam-nhin-da-phan-giai-cai-thien-kha-nang-nhin-cua-transformer-trong-xu-ly-hinh-anh-GAWVpZ0XJ05</guid>
            <description><![CDATA[Transformer, ban đầu được thiết kế cho xử lý ngôn ngữ tự nhiên (NLP), đã nhanh chóng trở thành một kiến trúc quan trọng trong thị giác máy tính. Các m...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Dương Xuân Bách</dc:creator>
            <pubDate>2025-01-13 00:53:44</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Scope and Applications of Computer Vision]]></title>
                        <link>https://viblo.asia/p/scope-and-applications-of-computer-vision-MkNLrQBbJgA</link>
            <guid isPermaLink="true">https://viblo.asia/p/scope-and-applications-of-computer-vision-MkNLrQBbJgA</guid>
            <description><![CDATA[Computer vision represents one of the multidisciplinary areas focused on enabling machines to understand visual information from the world. It is foun...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Gokul S</dc:creator>
            <pubDate>2024-09-30 13:07:54</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper reading] Towards Multi-Lingual Visual Question Answering]]></title>
                        <link>https://viblo.asia/p/paper-reading-towards-multi-lingual-visual-question-answering-0gdJzD3gVz5</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-reading-towards-multi-lingual-visual-question-answering-0gdJzD3gVz5</guid>
            <description><![CDATA[Giới thiệu chung

Visual Question Answering (VQA) là một bài toán thú vị mô phỏng gần nhất khả năng của con người. Nói một cách ngắn gọn, ta cần huấn ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Viblo AI</dc:creator>
            <pubDate>2024-06-07 17:00:00</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Góc nhìn tổng quan về bài toán Blind Image Super-Resolution]]></title>
                        <link>https://viblo.asia/p/goc-nhin-tong-quan-ve-bai-toan-blind-image-super-resolution-aNj4vxqKL6r</link>
            <guid isPermaLink="true">https://viblo.asia/p/goc-nhin-tong-quan-ve-bai-toan-blind-image-super-resolution-aNj4vxqKL6r</guid>
            <description><![CDATA[Blind image super-resolution (hay siêu phân giải mù), là bài toán nhằm siêu phân giải hình ảnh chất lượng thấp với degradation  chưa biết, đã thu hút ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hữu Đạt</dc:creator>
            <pubDate>2023-08-23 14:25:12</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[AI-Avatar] Tạo sinh Avatar cùng với Stable Diffusion]]></title>
                        <link>https://viblo.asia/p/ai-avatar-tao-sinh-avatar-cung-voi-stable-diffusion-pgjLNmoPJ32</link>
            <guid isPermaLink="true">https://viblo.asia/p/ai-avatar-tao-sinh-avatar-cung-voi-stable-diffusion-pgjLNmoPJ32</guid>
            <description><![CDATA[1. GIỚI THIỆU
Trong những năm gần đây, với sự phát triển của các mô hình Generative model các ứng dụng tự sinh dữ liệu text, video, ảnh càng trở nên n...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hoàng Anh Trịnh</dc:creator>
            <pubDate>2023-08-11 16:38:05</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Inpaint Anything - Công cụ chỉnh sửa ảnh bằng cách kết hợp SAM với Stable Diffusion]]></title>
                        <link>https://viblo.asia/p/inpaint-anything-cong-cu-chinh-sua-anh-bang-cach-ket-hop-sam-voi-stable-diffusion-Rk74axjkVeO</link>
            <guid isPermaLink="true">https://viblo.asia/p/inpaint-anything-cong-cu-chinh-sua-anh-bang-cach-ket-hop-sam-voi-stable-diffusion-Rk74axjkVeO</guid>
            <description><![CDATA[Mở đầu

Inpaint Anything được giới thiệu trong bài báo Inpaint Anything: Segment Anything Meets Image Inpainting (Yu et al., 2023) vào tháng 4 năm 202...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trần Đức Tuấn</dc:creator>
            <pubDate>2023-05-07 09:23:10</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Tổng quan về SAM - Công cụ segment vật thể bằng prompt cho bài toán Image Segmentation]]></title>
                        <link>https://viblo.asia/p/tong-quan-ve-sam-cong-cu-segment-vat-the-bang-prompt-cho-bai-toan-image-segmentation-Rk74axBDVeO</link>
            <guid isPermaLink="true">https://viblo.asia/p/tong-quan-ve-sam-cong-cu-segment-vat-the-bang-prompt-cho-bai-toan-image-segmentation-Rk74axBDVeO</guid>
            <description><![CDATA[Mở đầu
SAM (Segment Anything Model) được Meta AI công bố vào tháng 4 năm 2023 trong dự án "Segment Anything" với vai trò là một foundation model có th...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trần Đức Tuấn</dc:creator>
            <pubDate>2023-05-03 08:55:38</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] YOLOF: Lời tạm biệt cho multi-scale features trong Object Detection?]]></title>
                        <link>https://viblo.asia/p/paper-explain-yolof-loi-tam-biet-cho-multi-scale-features-trong-object-detection-5pPLkxMdVRZ</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-yolof-loi-tam-biet-cho-multi-scale-features-trong-object-detection-5pPLkxMdVRZ</guid>
            <description><![CDATA[Ôn lại kiến thức
Multi-scale features là gì và tại sao chúng lại cần thiết cho Object Detection?
Lấy ví dụ về một mạng Object Detection khá phổ biến: ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2023-05-05 16:17:36</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Clustering trong Computer Vision: Hướng đi mới thay thế CNN và Transformer?]]></title>
                        <link>https://viblo.asia/p/paper-explain-clustering-trong-computer-vision-huong-di-moi-thay-the-cnn-va-transformer-0gdJz7pvLz5</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-clustering-trong-computer-vision-huong-di-moi-thay-the-cnn-va-transformer-0gdJz7pvLz5</guid>
            <description><![CDATA[Tóm tắt
Ảnh là gì và làm thế nào để trích xuất features?
Convolutional Neural Network (CNN). CNN xem ảnh là các pixel có tổ chức theo dạng hình chữ nh...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2023-04-03 10:02:47</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Zero-shot Object Detection with Detic]]></title>
                        <link>https://viblo.asia/p/zero-shot-object-detection-with-detic-bXP4WAadL7G</link>
            <guid isPermaLink="true">https://viblo.asia/p/zero-shot-object-detection-with-detic-bXP4WAadL7G</guid>
            <description><![CDATA[![](https://images.viblo.asia/e43b0de5-e5ac-4a0b-836f-d10fd35c8b46.png)



Gần đây chúng ta đã quá quen thuộc với từ khóa "zero-shot", một trong những...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Thanh Luan Tran</dc:creator>
            <pubDate>2023-04-13 14:26:32</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Tính chất của Self-Attention và Transformer trong Computer Vision]]></title>
                        <link>https://viblo.asia/p/tinh-chat-cua-self-attention-va-transformer-trong-computer-vision-MkNLrO1bVgA</link>
            <guid isPermaLink="true">https://viblo.asia/p/tinh-chat-cua-self-attention-va-transformer-trong-computer-vision-MkNLrO1bVgA</guid>
            <description><![CDATA[Mở đầu
Qua 2 bài viết: cơ chế Attention trong Computer Vision và MetaFormer với cái tiêu đề đầy chế giễu, thì giống như mình là một hater của Self-Att...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2023-03-07 09:06:30</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Nhận diện và trích xuất thông tin biển số xe Việt Nam -  Identify and Extract Vietnamese License Plate information]]></title>
                        <link>https://viblo.asia/p/nhan-dien-va-trich-xuat-thong-tin-bien-so-xe-viet-nam-identify-and-extract-vietnamese-license-plate-information-5pPLkP08VRZ</link>
            <guid isPermaLink="true">https://viblo.asia/p/nhan-dien-va-trich-xuat-thong-tin-bien-so-xe-viet-nam-identify-and-extract-vietnamese-license-plate-information-5pPLkP08VRZ</guid>
            <description><![CDATA[1. Lời nói đầu
Cùng với sự phát triển về khoa học và tốc độ xử lý của máy tính ngày càng cao, việc áp dụng trí tuệ nhân tạo vào các lĩnh vực trong đời...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Leon Nguyen</dc:creator>
            <pubDate>2023-02-18 14:34:00</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Introduction Backpropagation CNN(Series 2)]]></title>
                        <link>https://viblo.asia/p/introduction-backpropagation-cnnseries-2-PwlVmj0145Z</link>
            <guid isPermaLink="true">https://viblo.asia/p/introduction-backpropagation-cnnseries-2-PwlVmj0145Z</guid>
            <description><![CDATA[Nhân dịp năm mới mình kính mong tất cả mọi người năm mới an lạnh thật nhiều niềm vui . Chúc mừng năm mới ngập tràn tiếng cười.
Phần 1: Backward CNN2D
...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hole Black Quyết</dc:creator>
            <pubDate>2023-02-17 20:16:13</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] RTMDet: YOLO của OpenMMLab]]></title>
                        <link>https://viblo.asia/p/paper-explain-rtmdet-yolo-cua-openmmlab-n1j4lKjaVwl</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-rtmdet-yolo-cua-openmmlab-n1j4lKjaVwl</guid>
            <description><![CDATA[Mở đầu
Mình khá là thích OpenMMLab, một team nghiên cứu đã cung cấp rất nhiều repo tăng tốc các thử nghiệm như MMDetection, MMSegmentation, MMCV,... Đ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2023-05-04 16:06:54</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Phân loại hình ảnh với Vision Transformer]]></title>
                        <link>https://viblo.asia/p/phan-loai-hinh-anh-voi-vision-transformer-5OXLAowk4Gr</link>
            <guid isPermaLink="true">https://viblo.asia/p/phan-loai-hinh-anh-voi-vision-transformer-5OXLAowk4Gr</guid>
            <description><![CDATA[Ví dụ này triển khai mô hình Vision Transformer (ViT) của Alexey Dosovitskiy để phân loại hình ảnh và thể hiện mô hình đó trên tập dữ liệu CIFAR-100. ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Chung Hoàng Ngọc Kỳ An</dc:creator>
            <pubDate>2023-01-19 10:17:44</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] MetaFormer: Khi Attention is NOT all you need cho bài toán phân loại ảnh]]></title>
                        <link>https://viblo.asia/p/paper-explain-metaformer-khi-attention-is-not-all-you-need-cho-bai-toan-phan-loai-anh-zXRJ8n1ZVGq</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-metaformer-khi-attention-is-not-all-you-need-cho-bai-toan-phan-loai-anh-zXRJ8n1ZVGq</guid>
            <description><![CDATA[Yêu cầu nhỏ
Hiểu về các lớp Norm khác nhau hoạt động như nào: BatchNorm (BN), GroupNorm (GN), LayerNorm (LN) và biết cách sử dụng Pytorch

Hình 0. Các...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2023-01-31 14:01:59</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] TOOD: Thống nhất Classification và Localization cho Object Detection]]></title>
                        <link>https://viblo.asia/p/paper-explain-tood-thong-nhat-classification-va-localization-cho-object-detection-qPoL7XK1Vvk</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-tood-thong-nhat-classification-va-localization-cho-object-detection-qPoL7XK1Vvk</guid>
            <description><![CDATA[Tại sao lại là TOOD?
- Vì trong paper này có một phần mà các paper về Object Detection hiện nay áp dụng khá nhiều, tiêu biểu là YOLOv6 hay DAMO-YOLO m...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2022-12-13 11:03:33</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[YOLOv4 - Kỷ nguyên mới cho những mô hình họ YOLO]]></title>
                        <link>https://viblo.asia/p/yolov4-ky-nguyen-moi-cho-nhung-mo-hinh-ho-yolo-vlZL9N0dVQK</link>
            <guid isPermaLink="true">https://viblo.asia/p/yolov4-ky-nguyen-moi-cho-nhung-mo-hinh-ho-yolo-vlZL9N0dVQK</guid>
            <description><![CDATA[
Trong lĩnh vực Computer Vision, bài toán Object Detection là một bài toán hết sức phổ biến. Như cái tên gọi của nó, mục tiêu của bài toán này là phát...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Vũ</dc:creator>
            <pubDate>2023-03-09 08:17:43</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Rep-Optimizer: Re-params Optimizer thay vì Re-params model của bạn]]></title>
                        <link>https://viblo.asia/p/paper-explain-rep-optimizer-re-params-optimizer-thay-vi-re-params-model-cua-ban-AZoJj7D2LY7</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-rep-optimizer-re-params-optimizer-thay-vi-re-params-model-cua-ban-AZoJj7D2LY7</guid>
            <description><![CDATA[Một số khái niệm cần nắm được
- Re-parameterize: Là kĩ thuật thay đổi parameters của một layers (kernel của một lớp Conv) theo dạng biểu diễn khác. Về...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2022-11-23 14:57:54</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Imagen - Mô hình SOTA giải quyết bài toán Text-to-Image]]></title>
                        <link>https://viblo.asia/p/imagen-mo-hinh-sota-giai-quyet-bai-toan-text-to-image-gwd43BrAVX9</link>
            <guid isPermaLink="true">https://viblo.asia/p/imagen-mo-hinh-sota-giai-quyet-bai-toan-text-to-image-gwd43BrAVX9</guid>
            <description><![CDATA[
Imagen - mô hình mới được công bố gần đây bởi Google với khả năng generate hình ảnh từ đoạn text mô tả bất kỳ, cho dù ảnh đó không có thật hoặc phi l...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Quốc Hưng Vũ</dc:creator>
            <pubDate>2022-10-24 23:44:47</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Một chút về cơ chế Attention trong Computer Vision]]></title>
                        <link>https://viblo.asia/p/mot-chut-ve-co-che-attention-trong-computer-vision-x7Z4D622LnX</link>
            <guid isPermaLink="true">https://viblo.asia/p/mot-chut-ve-co-che-attention-trong-computer-vision-x7Z4D622LnX</guid>
            <description><![CDATA[Mở đầu
Cơ chế Attention (Attention mechanism) là một cơ chế vô cùng hay và nhận được rất nhiều sự phát triển gần đây. Có những model được tạo thành lấ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2022-10-14 17:11:52</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Giới thiệu về Diffusion model (Series 1)]]></title>
                        <link>https://viblo.asia/p/gioi-thieu-ve-diffusion-model-series-1-EvbLbOKWVnk</link>
            <guid isPermaLink="true">https://viblo.asia/p/gioi-thieu-ve-diffusion-model-series-1-EvbLbOKWVnk</guid>
            <description><![CDATA[1 . Giới thiệu về sơ lược về diffusion model
Trong bài viết này mình sẽ giới thiệu sơ qua về mô hình và ứng dụng của chúng quan trọng hơn hết với nhữn...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hole Black Quyết</dc:creator>
            <pubDate>2022-09-16 17:26:04</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[YOLOv3 - tinh chỉnh để trở nên tốt hơn]]></title>
                        <link>https://viblo.asia/p/yolov3-tinh-chinh-de-tro-nen-tot-hon-E1XVO3kPVMz</link>
            <guid isPermaLink="true">https://viblo.asia/p/yolov3-tinh-chinh-de-tro-nen-tot-hon-E1XVO3kPVMz</guid>
            <description><![CDATA[1. Giới thiệu
Tiếp tục về YOLO series sẽ là sự giới thiệu và trình bày về YOLOv3. YOLOv3 đưa ra những sự điều chỉnh nhỏ so với phiên bản trước đó, khô...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hao Ng</dc:creator>
            <pubDate>2022-11-03 08:53:56</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[YOLOv2: Tốt hơn, nhanh hơn và mạnh mẽ hơn]]></title>
                        <link>https://viblo.asia/p/yolov2-tot-hon-nhanh-hon-va-manh-me-hon-oK9Vy0AXLQR</link>
            <guid isPermaLink="true">https://viblo.asia/p/yolov2-tot-hon-nhanh-hon-va-manh-me-hon-oK9Vy0AXLQR</guid>
            <description><![CDATA[1. Giới thiệu
Đúng như tiêu đề ấn tượng của bài báo: "YOLO9000: Better, Faster, Stronger", YOLOv2 kế thừa và phát triển tiếp từ YOLOv1 với hàng loạt n...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hao Ng</dc:creator>
            <pubDate>2022-09-13 11:00:38</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[YOLOv1 - Nhận diện vật thể chỉ ngó 1 lần - Khởi nguồn]]></title>
                        <link>https://viblo.asia/p/yolov1-nhan-dien-vat-the-chi-ngo-1-lan-khoi-nguon-aNj4v0b0L6r</link>
            <guid isPermaLink="true">https://viblo.asia/p/yolov1-nhan-dien-vat-the-chi-ngo-1-lan-khoi-nguon-aNj4v0b0L6r</guid>
            <description><![CDATA[1. Giới thiệu
Ở thời điểm YOLO được ra mắt bởi Facebook AI Research (FAIR), các công trình trước đó về Object Detection đều sử dụng các bộ phân loại (...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hao Ng</dc:creator>
            <pubDate>2022-09-07 17:10:42</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain][End-to-End Dense Video Captioning with Parallel Decoding] PDVC -  Hướng tiếp cận end-to-end với giải mã song song cho bài toán Dense Video Captioning]]></title>
                        <link>https://viblo.asia/p/paper-explainend-to-end-dense-video-captioning-with-parallel-decoding-pdvc-huong-tiep-can-end-to-end-voi-giai-ma-song-song-cho-bai-toan-dense-video-captioning-gDVK2PaAlLj</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explainend-to-end-dense-video-captioning-with-parallel-decoding-pdvc-huong-tiep-can-end-to-end-voi-giai-ma-song-song-cho-bai-toan-dense-video-captioning-gDVK2PaAlLj</guid>
            <description><![CDATA[Xin chào các bạn. Chắc hẳn mọi người cũng đã từng nghe qua đến cái tên Video Understanding với rất nhiều các tác vụ con kèm theo đó là khá model nổi t...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hữu Đạt</dc:creator>
            <pubDate>2022-06-16 18:03:35</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Các chỉ số đánh giá được sử dụng cho bài toán Image Generation: IS, FID, PSNR, SSIM,...]]></title>
                        <link>https://viblo.asia/p/cac-chi-so-danh-gia-duoc-su-dung-cho-bai-toan-image-generation-is-fid-psnr-ssim-3P0lPJXPKox</link>
            <guid isPermaLink="true">https://viblo.asia/p/cac-chi-so-danh-gia-duoc-su-dung-cho-bai-toan-image-generation-is-fid-psnr-ssim-3P0lPJXPKox</guid>
            <description><![CDATA[1. Giới thiệu về bài toán Image Generation

   Bài toán Image Generation là một trong những bài toán cũng đang được quan tâm rất nhiều trong thời gian...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hữu Đạt</dc:creator>
            <pubDate>2022-06-11 10:26:57</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Learning Temporally Invariant and Localizable Features via Data Augmentation for Video Recognition - Bàn luận 1 chút về video augmentation.]]></title>
                        <link>https://viblo.asia/p/paper-explain-learning-temporally-invariant-and-localizable-features-via-data-augmentation-for-video-recognition-ban-luan-1-chut-ve-video-augmentation-RnB5pAL6KPG</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-learning-temporally-invariant-and-localizable-features-via-data-augmentation-for-video-recognition-ban-luan-1-chut-ve-video-augmentation-RnB5pAL6KPG</guid>
            <description><![CDATA[Tiếp tục series Paper Explain đang dang dở về topic Action Recognition, trong bài viết này, mình muốn bàn một chút về data: Data Augmentation.

Chắc m...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bui Tien Tung</dc:creator>
            <pubDate>2022-05-31 23:56:46</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Revisiting Skeleton-based Action Recognition - BKAI-NAVER Challenge 2022 Top1 Solution Baseline]]></title>
                        <link>https://viblo.asia/p/paper-explain-revisiting-skeleton-based-action-recognition-bkai-naver-challenge-2022-top1-solution-baseline-naQZRRzXZvx</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-revisiting-skeleton-based-action-recognition-bkai-naver-challenge-2022-top1-solution-baseline-naQZRRzXZvx</guid>
            <description><![CDATA[Tiêu đề hơi giật tít một chút, nhưng gần đây, mình cùng team có tham gia một challenge được tổ chức bởi trung tâm nghiên cứu BKAI kết hợp với tập đoàn...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bui Tien Tung</dc:creator>
            <pubDate>2022-05-30 02:07:12</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] DualStyleGAN:  Exemplar-Based High-Resolution Portrait Style Transfer]]></title>
                        <link>https://viblo.asia/p/paper-explain-dualstylegan-exemplar-based-high-resolution-portrait-style-transfer-Do754rRLZM6</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-dualstylegan-exemplar-based-high-resolution-portrait-style-transfer-Do754rRLZM6</guid>
            <description><![CDATA[
Introduction
Gần đây, nhiều nghiên cứu chỉ ra rằng StyleGAN có thể thực hiện style transfer chất lượng cao chỉ với một lượng dữ liệu hạn chế bằng một...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hieu Bui</dc:creator>
            <pubDate>2022-05-29 20:09:41</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain]Sinh ảnh cùng với MixNMatch: độ chân thực đến đáng gờm]]></title>
                        <link>https://viblo.asia/p/paper-explainsinh-anh-cung-voi-mixnmatch-do-chan-thuc-den-dang-gom-OeVKBrjdKkW</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explainsinh-anh-cung-voi-mixnmatch-do-chan-thuc-den-dang-gom-OeVKBrjdKkW</guid>
            <description><![CDATA[Tổng quan

Dạo một vòng mấy trang chia sẻ kiên thức để chống lười, đập vào mắt mình một bài viết với tiêu đề khá giật tít, mà kiểu nó sẽ áp dụng được ...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trung Đức</dc:creator>
            <pubDate>2022-05-12 14:21:12</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[NeuronNetwork Bài toán Traffic Sign Detection với bộ dữ liệu TT100K]]></title>
                        <link>https://viblo.asia/p/neuronnetwork-bai-toan-traffic-sign-detection-voi-bo-du-lieu-tt100k-3P0lP83blox</link>
            <guid isPermaLink="true">https://viblo.asia/p/neuronnetwork-bai-toan-traffic-sign-detection-voi-bo-du-lieu-tt100k-3P0lP83blox</guid>
            <description><![CDATA[Trích dẫn:
Nội dung bài viết phân tích và tổng hợp các kiến thức có trong paper ["Traffic-Sign Detection and Classification in the Wild"] (2016)(https...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trần Đình Lâm</dc:creator>
            <pubDate>2022-05-02 21:28:16</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection]]></title>
                        <link>https://viblo.asia/p/paper-explain-generalized-focal-loss-learning-qualified-and-distributed-bounding-boxes-for-dense-object-detection-bWrZnAwmKxw</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-generalized-focal-loss-learning-qualified-and-distributed-bounding-boxes-for-dense-object-detection-bWrZnAwmKxw</guid>
            <description><![CDATA[Khái niệm cần biết
- Dense Object Detector: Một mạng Object Detection thực hiện lấy mẫu các vị trí có khả năng tồn tại Object trong ảnh một cách cực d...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nguyen Mai</dc:creator>
            <pubDate>2022-04-12 13:39:03</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[[Paper Explain] Contrastive Learning for Label-Efficient Semantic Segmentation]]></title>
                        <link>https://viblo.asia/p/paper-explain-contrastive-learning-for-label-efficient-semantic-segmentation-Qbq5QRrLKD8</link>
            <guid isPermaLink="true">https://viblo.asia/p/paper-explain-contrastive-learning-for-label-efficient-semantic-segmentation-Qbq5QRrLKD8</guid>
            <description><![CDATA[Lời mở đầu
Tiếp nối việc phân tích paper, hôm nay mình sẽ cùng các bạn phân tích 1 paper liên quan dến bài toán Semantic Segmentation và phương pháp C...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Trung Đức</dc:creator>
            <pubDate>2022-03-07 16:45:05</pubDate>
                                                                                                        </item>
                <item>
            <title><![CDATA[Giới thiệu về Variational Autoencoder]]></title>
                        <link>https://viblo.asia/p/gioi-thieu-ve-variational-autoencoder-Eb85ozN0l2G</link>
            <guid isPermaLink="true">https://viblo.asia/p/gioi-thieu-ve-variational-autoencoder-Eb85ozN0l2G</guid>
            <description><![CDATA[Introduction
Xin chào mọi người, trong bài viết ngày hôm này minh sẽ cùng mọi người tìm hiểu về  Variational Autoencoder (VAE), một loại generative mo...]]></description>
                        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Hieu Bui</dc:creator>
            <pubDate>2022-02-17 17:04:35</pubDate>
                                                                                                        </item>
            </channel>
</rss>
