DeepGlyph
DeepGlyph ã¯ã€ãƒ•ã‚©ãƒ³ãƒˆåˆ¶ä½œã®ãŸã‚ã®æ–°ã—ã„アプリケーションã§ã™ã€‚ 深層å¦ç¿’ã®æŠ€è¡“を使ã„ã€æ•°æ–‡å—ã®ã‚µãƒ³ãƒ—ルã‹ã‚‰ã‚ãªãŸã ã‘ã®ã‚ªãƒªã‚¸ãƒŠãƒ«ãª Web フォントを生æˆã—ã¾ã™ã€‚ DeepGlyph ã¯ã€ãƒ•ã‚©ãƒ³ãƒˆåˆ¶ä½œã®ãŸã‚ã®ãƒ™ã‚¯ã‚¿ãƒ¼ç·¨é›†æ©Ÿèƒ½ã‚’æ載。 自動生æˆã§ã¯å†ç¾ã§ããªã‹ã£ãŸç®‡æ‰€ã‚’調整ã—ã¦ã€ç†æƒ³ã®ãƒ•ã‚©ãƒ³ãƒˆã‚’仕上ã’ã¾ã—ょã†ã€‚
One-stage object detection
Object detection is the computer vision technique for finding objects of interest in an image: This is more advanced than classification, which only tells you what the “main subject†of the image is — whereas object detection can find multiple objects, classify them, and locate where they are in the image. An object detection model predicts bounding boxes, one for each object it finds, as well as
長文を３行ãらã„ã§çºã‚ã¦ãれるエンジン IMAKITAを作ã£ã¦ã¿ã¾ã—㟠- コンピュータ将棋 Qhapaq
ã¾ã•ã‹ã®Gigazineデビューをã—ã¾ã—ãŸã€‚ã‚ã‚ŠãŒã¨ã†ã”ã–ã„ã¾ã™ï¼ï¼ gigazine.net å”çªã§ã™ãŒçš†æ§˜ã¯å‰ã„人ã®é•·è©±ã«è‹›ã€…ã—ãŸã“ã¨ã¯ãªã„ã§ã—ょã†ã‹ã€‚言ã„ãŸã„ã“ã¨ã¯çŸã„ã®ã«æžè‘‰ã‚’ã¤ã‘ãŸé•·æ–‡ã‚’é€ã‚‰ã‚Œã‚‹ã®ã«ã‚¦ãƒ³ã‚¶ãƒªã—ãŸã“ã¨ã¯ãªã„ã§ã—ょã†ã‹ã€‚ ãã‚“ãªçš†æ§˜ã®å£°ï¼ˆï¼Ÿï¼‰ã‚’å—ã‘ã¦ã€é•·æ–‡ã‚’３行ãらã„ã§çºã‚ã¦ãれる（厳密ã«ã¯ã€æ–‡ç« 全体ã®ä¸ã§ç‰¹ã«é‡è¦åº¦ã®é«˜ã„文を抽出ã—ã¦ãれる）エンジン IMAKITAを作ã£ã¦ã¿ã¾ã—ãŸã€‚ https://www.qhapaq.org/imakita/ 使ã„方： ・テã‚ストボックスã«æ–‡ç« を入れる（日本語ã¯ã€Œã€‚ã€/「ï¼ã€/「？ã€ã€ä¸å›½èªžã¯ã€Œã€‚ã€åŒºåˆ‡ã‚Šã€è‹±èªžã€ã‚¹ãƒšã‚¤ãƒ³èªžã€ãƒ‰ã‚¤ãƒ„語ã€ãƒ•ãƒ©ãƒ³ã‚¹èªžã€ãƒãƒ«ãƒˆã‚¬ãƒ«èªžã€ã‚¤ã‚¿ãƒªã‚¢èªžã¯ã€Œ.ã€åŒºåˆ‡ã‚Šã€‚日本語ã®ã¿æ”¹è¡Œã‚‚区切る機能を試験的ã«å°Žå…¥ä¸ï¼‰ ・Squeezeボタンを押㙠・çµæžœã‚’楽ã—ã‚€ 使用上ã®æ³¨æ„： ・無ä¿è¨¼ã§ã™ ãƒ»æ–‡ç« ãŒé•·ã™ãŽã‚‹ã¨è½ã¡ã¾ã™
How to easily do Object Detection on Drone Imagery using Deep learning
Section 1: Aerial Imagery — a brief backgroundMan has always been fascinated with a view of the world from the top — building watch-towers, high fortwalls, capturing the highest mountain peak. To capture a glimpse and share it with the world, people went to great lengths to defy gravity, enlisting the help of ladders, tall buildings, kites, balloons, planes, and rockets. Images of San Francisco ta
ç”»åƒã‚ャプションã¨å‹•ä½œèªè˜ã®æœ€å‰ç·š 〜データセットã«æ³¨ç›®ã—ã¦ã€œï¼ˆç¬¬17回ステアラボ人工知能セミナー）
ç”»åƒã‚ャプションã¨å‹•ä½œèªè˜ã®æœ€å‰ç·š 〜データセットã«æ³¨ç›®ã—ã¦ã€œï¼ˆç¬¬17回ステアラボ人工知能セミナー）AI-enhanced description This document summarizes several datasets for image captioning, video classification, action recognition, and temporal localization. It describes the purpose, collection process, annotation format, examples and references for datasets including MS COCO, Visual Genome, Flickr8K/30K, Kinetics, Charades, AVA, STAIR Captions and
Actcast紹介資料 / actcast-ja
Idein Inc.ãŒé–‹ç™ºä¸ã®ã‚¨ãƒƒã‚¸ã‚³ãƒ³ãƒ”ューティングプラットフォームサービスActcastã®ç´¹ä»‹è³‡æ–™ã§ã™ã€‚ 様々ãªå®Ÿä¸–ç•Œæƒ…å ±ã¨Webサービスã¨ãŒé€£æºã™ã‚‹å…ˆé€²çš„ãªIoTシステムを手軽ã«å®‰ä¾¡ã«æ§‹ç¯‰ã™ã‚‹äº‹ãŒã§ãã¾ã™ã€‚ ディープラーニングモデルã«ã‚ˆã‚‹æŽ¨è«–ã¯ã‚¨ãƒƒã‚¸ãƒ‡ãƒã‚¤ã‚¹ä¸Šã§å®Ÿè¡Œã•ã‚Œã€ã‚³ã‚¹ãƒˆã®æœ€é©åŒ–やプライ…
YOLO: Real-Time Object Detection
How It Works Prior detection systems repurpose classifiers or localizers to perform detection. They apply the model to an image at multiple locations and scales. High scoring regions of the image are considered detections. We use a totally different approach. We apply a single neural network to the full image. This network divides the image into regions and predicts bounding boxes and probabilitie
YOLOv2ã®ãƒªã‚¢ãƒ«ã‚¿ã‚¤ãƒ 物体検出をTensorFlowã¨Pythonã§å®Ÿè£…ã™ã‚‹æ–¹æ³•
YOLO v2ã‚’ã©ã†ã—ã¦ã‚‚Pythonã§ä½¿ã£ã¦ã¿ãŸã‹ã£ãŸã®ã§ä½œã£ã¦ã¿ãŸ ã“ã‚“ã«ã¡ã¯ã€‚ AI coordinatorã®æ¸…水秀樹ã§ã™ã€‚ 数多ãã‚るオブジェクト物体検出ã®ä¸ã§ã€å‡¦ç†é€Ÿåº¦ãŒæœ€ã‚‚æ—©ã„？ã¨è¨€ã‚ã‚Œã¦ã„ã‚‹YOLO v2を試ã—ã¦ã¿ã¾ã—ãŸã€‚ å…¬å¼ã‚µã‚¤ãƒˆã®é€šã‚Šã‚„ã£ã¦ã€ç’°å¢ƒã®ã‚»ãƒƒãƒˆã‚¢ãƒƒãƒ—ã¨é™æ¢ç”»ã®ã‚ªãƒ–ジェクト物体検出をæç”»ã§ãã‚‹ã¨ã“ã‚ã¾ã§ã¯ç°¡å˜ã«ã§ãã¾ã—ãŸãŒã€ãªãœã‹å‹•ç”»ã«ãªã‚‹ã¨ã‚¨ãƒ©ãƒ¼ã«ãªã£ã¦ã—ã¾ã„ã¾ã—ãŸã€‚ ãã‚“ãªã‚ã‘ã§ã€å…¬å¼ã‚µã‚¤ãƒˆã®ã‚„ã‚Šæ–¹ã§ã¯ã‚¤ãƒžã‚¤ãƒä¸è‰¯æ¶ˆåŒ–ã®ã¾ã¾å…ˆã«é€²ã‚ãªããªã£ã¦ã—ã¾ã£ãŸã®ã§ã€å°‘ã—æ€è€ƒã‚’変ãˆã€é™æ¢ç”»ã§ã‚ªãƒ–ジェクト物体検出ãŒã§ãã‚‹ãªã‚‰ã€Pythonã§ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã‚’組ã¿ç›´ã›ã°ã€ãƒªã‚¢ãƒ«ã‚¿ã‚¤ãƒ æ˜ åƒã§ã‚‚YOLOを試ã›ã‚‹ã¨è€ƒãˆã€å®Ÿéš›ã«Pythonã§å‹•ã‹ã›ã‚‹ã‚ˆã†ã«ä½œæˆã—ã¦ã¿ã¾ã—ãŸã€‚ ä¸ã€…苦労ã—ã¾ã—ãŸãŒã€ãªã‚“ã¨ã‹å®Ÿè£…ã§ããŸã®ã§ã€ãã®å†…容を紹介ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ å‚考ã«ã•ã›ã¦é ‚ã„ãŸã‚µã‚¤ãƒˆã®ç´¹ä»‹ ã¡
Image Completion with Deep Learning in TensorFlow
Introduction Content-aware fill is a powerful tool designers and photographers use to fill in unwanted or missing parts of images. Image completion and inpainting are closely related technologies used to fill in missing or corrupted parts of images. There are many ways to do content-aware fill, image completion, and inpainting. In this blog post, I present Raymond Yeh and Chen Chen et al.’s paper
Colorful Image Colorization
Example input grayscale photos and output colorizations from our algorithm. These examples are cases where our model works especially well. For randomly selected examples, see the Performance comparisons section below. Welcome! Computer vision algorithms often work well on some images, but fail on others. Ours is like this too. We believe our work is a significant step forward in solving the color
GitHub - kjw0612/awesome-deep-vision: A curated list of deep learning resources for computer vision
(from Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton, ImageNet Classification with Deep Convolutional Neural Networks, NIPS, 2012.) Microsoft (Deep Residual Learning) [Paper][Slide] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun, Deep Residual Learning for Image Recognition, arXiv:1512.03385. Microsoft (PReLu/Weight Initialization) [Paper] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun,
Faster Deep Learning with Theano & GPUs
Build and train neural networks in Python.Using the GPU, I'll show that we can train deep belief networks up to 15x faster than using just the CPU, cutting training time down from hours to minutes. Why are GPUs useful?When you think of high-performance graphics cards, data science may not be the first thing that comes to mind. However, computer graphics and data science have one important thing in
ディープラーニングを使ã£ãŸã‚¤ãƒ¡ãƒ¼ã‚¸ã®åˆ‡ã‚ŠæŠœã | カメリオ開発者ブãƒã‚°
ã“ã‚“ã«ã¡ã¯ã€ã‚·ãƒã‚¿ã‚¢ã‚ラã§ã™ã€‚ã“ã®åº¦PyDataã®æœ¬å®¶ã§ã‚るアメリカã®ã‚³ãƒŸãƒ¥ãƒ‹ãƒ†ã‚£ãƒ¼ã§åŠå¹´ã«ä¸€åº¦é–‹å‚¬ã•ã‚Œã¦ã„ã‚‹PyDataカンファレンスã«å‡ºå¸ã™ã‚‹ãŸã‚ã€NYCã«è¡Œã£ã¦æ¥ã¾ã—ãŸã€‚11/22-11/23ã®äºŒæ—¥é–“ã®æ—¥ç¨‹ã§è¡Œã‚ã‚Œã€å»¶ã¹250人ã»ã©ãŒå‚åŠ ã—ãŸã‚¤ãƒ™ãƒ³ãƒˆã§ã™ã€‚ãã®æ™‚ã®æ¨¡æ§˜ã¯ã€å…ˆæ—¥ã®PyData Tokyo第二回ミートアップã§ã‚‚ã”説明ã•ã›ã¦ã„ãŸã ãã€ã¾ãŸå¾Œæ—¥è¨˜äº‹åŒ–ã•ã‚Œã‚‹ã¨æ€ã„ã¾ã™ã®ã§ã€ãã¡ã‚‰ã‚’ãœã²ã”覧ã„ãŸã ã‘ã‚Œã°ã¨æ€ã„ã¾ã™ã€‚ 今回ã¯ãã®PyData NYCカンファレンスã§ç§ãŒç™ºè¡¨ã—ã¦ããŸãƒŸãƒ‹ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã«ã¤ã„ã¦ãŠè©±ã—ã¾ã™ã€‚最近å„所ã§è©±é¡Œã«ä¸ŠãŒã‚‹ãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã§ã™ãŒã€ã“れを使ã£ãŸå¿œç”¨ã‚’「カメリオã€ã®ã‚µãƒ¼ãƒ“スå‘上ã®ãŸã‚ã«ä½¿ãˆãªã„ã‹ã€ã¨ã„ã†ã®ãŒãã‚‚ãã‚‚ã®ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã®ç€æƒ³ã§ã—ãŸã€‚今回PyData Tokyoオーガナイザーã¨ã—ã¦ã€ã¾ãŸãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã§è‰²ã€…ã¨é¢ç™½ã„実験をã—ã¦ã„ã‚‹ç”°ä¸ã•ã‚“（@a
Introducing: Flickr PARK or BIRD | code.flickr.com
tl;dr: Check it out at parkorbird.flickr.com! We at Flickr are not ones to back down from a challenge. Especially when that challenge comes in webcomic form. And especially when that webcomic is xkcd. So, when we saw this xkcd comic we thought, “we’ve got to do thatâ€: In fact, we already had the technology in place to do these things.  Like the woman in the comic says, determining whether a photo
Deep Learning for Image Recognition in Python
Deep Learning for Image Recognition in PythonAI-enhanced description The document discusses the application of deep learning for image recognition in Python, specifically focusing on distinguishing between images such as dogs and cats. It highlights the use of pre-trained networks and compares accuracy rates between traditional methods and deep learning, with the latter achieving over 95% accuracy
実装ディープラーニング
ã¯ã˜ãƒ‘ã‚¿LT資料ã§ã™ã€‚ディープラーニングã®ãƒ©ã‚¤ãƒ–ラリã®ä½¿ã„æ–¹ãŒçŸ¥ã‚ŠãŸã„æ–¹ã¯15ページã‹ã‚‰ã”覧ãã ã•ã„。
1
ランã‚ング
ランã‚ング
ãŠçŸ¥ã‚‰ã›
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
http://blog.alexandrecarlier.com/deepsvg/
GitHub - thoughtworksarts/EmoPy: A deep neural net toolkit for emotion analysis via Facial Expression Recognition (FER)
Chainer: A flexible framework for neural networks
{{#tags}}- {{label}}
{{/tags}}