{"id":740,"date":"2024-12-16T15:07:16","date_gmt":"2024-12-16T20:07:16","guid":{"rendered":"https:\/\/waterloobae.com\/?p=740"},"modified":"2025-08-20T14:23:16","modified_gmt":"2025-08-20T18:23:16","slug":"reading-crowdmark-cover-page-with-locally-installed-llama-3-2-vision-11b","status":"publish","type":"post","link":"https:\/\/waterloobae.com\/blog\/2024\/12\/16\/reading-crowdmark-cover-page-with-locally-installed-llama-3-2-vision-11b\/","title":{"rendered":"[Crowdmark] Reading Crowdmark cover page with locally installed Llama 3.2 Vision 11B"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">To experiment how to collect student information when student rooster is not predefined, reading Crowdmark cover page with locally installed Llama 3.2 Vision 11B was attempted to evaluate the accuracy of OCR function from the model. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The image is one of typical Crowdmark cover page with QR code and the area for the student first and last name, and student id, which is worded as participant ID below. Most of information is blurred since it is a real cover page for that contest. So, you may not be able to see how bad or good student&#8217;s handwriting was.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1018\" height=\"1024\" src=\"https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1018x1024.png\" alt=\"\" class=\"wp-image-742\" style=\"width:497px;height:auto\" srcset=\"https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1018x1024.png 1018w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-298x300.png 298w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-150x150.png 150w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-768x773.png 768w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image.png 1182w\" sizes=\"auto, (max-width: 1018px) 100vw, 1018px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The prompt was requesting specific information from the cover page and return them in JSON format as below. Those provided information were accurate. Since this model was running in my local machine with Ollama, it took around 13 seconds to get response. The results are returned as below. <\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"612\" src=\"https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1-1024x612.png\" alt=\"\" class=\"wp-image-743\" style=\"width:624px;height:auto\" srcset=\"https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1-1024x612.png 1024w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1-300x179.png 300w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1-768x459.png 768w, https:\/\/waterloobae.com\/wp-content\/uploads\/2024\/12\/image-1.png 1103w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Python source codes used can be found below.<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code lang=\"python\" class=\"language-python\">import ollama\nimport datetime\n\n# Get the current date and time\ncurrent_datetime = datetime.datetime.now()\n# Print the current date and time\nprint(current_datetime)\n\nwith open('CoverPage2.jpg', 'rb') as file:\n  response = ollama.chat(\n    model='x\/llama3.2-vision:latest',\n    messages=[\n      {\n        'role': 'user',\n        'content': '''Can you read First name, Last name, \n        Participant ID, Email address, School, City\/Town, and\n        Country. Then, give me information in JSON Format''',\n        'images': [file.read()],\n      },\n    ],\n  )\n\nprint(response['message']['content'])\n\n# Get the current date and time\ncurrent_datetime = datetime.datetime.now()\n# Print the current date and time\nprint(current_datetime)<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"<p>To experiment how to collect student information when student rooster is not predefined, reading Crowdmark cover page with locally installed Llama 3.2 Vision 11B was attempted to evaluate the accuracy of OCR function from the model. The image is one of typical Crowdmark cover page with QR code and the area for the student first [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[16,4],"tags":[],"class_list":["post-740","post","type-post","status-publish","format-standard","hentry","category-crowdmark","category-it-tips"],"_links":{"self":[{"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/posts\/740","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/comments?post=740"}],"version-history":[{"count":3,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/posts\/740\/revisions"}],"predecessor-version":[{"id":787,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/posts\/740\/revisions\/787"}],"wp:attachment":[{"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/media?parent=740"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/categories?post=740"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/waterloobae.com\/blog\/wp-json\/wp\/v2\/tags?post=740"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}