--- title: TAPe+ML: detection, classification, and now segmentation... description: TAPe+ML is a compact computer vision model (~0.1M parameters) that simultaneously handles object detection, classification, and instance segmentation in a... canonical: https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: TAPe+ML: detection, classification, and now segmentation of the SOTA‑level in 0.1M parameters | daily.dev og:description: TAPe+ML is a compact computer vision model (~0.1M parameters) that simultaneously handles object detection, classification, and instance segmentation in a... og:url: https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee og:image: https://api.daily.dev/og/posts/4e2gAChee.png og:image:alt: TAPe+ML: detection, classification, and now segmentation of the SOTA‑level in 0.1M parameters og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # TAPe+ML: detection, classification, and now segmentation of the SOTA‑level in 0.1M parameters **[Medium](https://daily.dev/sources/medium_js)** · 11 min read · 0 upvotes · 0 comments ## Summary TAPe+ML is a compact computer vision model (~0.1M parameters) that simultaneously handles object detection, classification, and instance segmentation in a single unified architecture. The segmentation module builds pixel-level contours through a three-stage process (boundary detection, contour closure, surface merging) without increasing parameter count or requiring separate training data. Trained on LVIS masks and panoptic COCO negatives, the model achieves 97–98% object coverage at AP@75 on LVIS and outperforms RF-DETR-Seg-2XL and YOLO26-X-Seg on COCO Mask AP metrics, while running the full pipeline in ~15ms on a GTX 1070Ti. Detection and segmentation share a single joint loss, so each task mutually improves the other during training. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [SAM 3: Concept-Based Visual Understanding and Segmentation](https://daily.dev/posts/sam-3-concept-based-visual-understanding-and-segmentation-gwnb1xefs) · PyImageSearch · 2 upvotes · 0 comments - [Fine-Tuning SOTA Object Detection Models on Real-World Datasets](https://daily.dev/posts/fine-tuning-sota-object-detection-models-on-real-world-datasets-jumlvrmgq) · JetBrains · 0 upvotes · 0 comments - [Falcon Perception](https://daily.dev/posts/falcon-perception-x4ei6gikw) · Hugging Face · 0 upvotes · 0 comments - [Researchers Solved a Decade-old Problem in Object Detection](https://daily.dev/posts/researchers-solved-a-decade-old-problem-in-object-detection-nr8nvuooe) · Daily Dose of Data Science \| Avi Chawla \| Substack · 2 upvotes · 0 comments --- Tags: [#machine-learning](https://daily.dev/tags/machine-learning), [#deep-learning](https://daily.dev/tags/deep-learning), [#computer-vision](https://daily.dev/tags/computer-vision), [#object-detection](https://daily.dev/tags/object-detection) [View this post on daily.dev](https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"TAPe+ML: detection, classification, and now segmentation of the SOTA‑level in 0.1M parameters","url":"https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee"},"datePublished":"2026-07-31T18:35:06.609Z","dateModified":"2026-07-31T18:36:26.598Z","description":"TAPe+ML is a compact computer vision model (~0.1M parameters) that simultaneously handles object detection, classification, and instance segmentation in a...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4b4c68374d7dd1f3edc86cf1b8f5e53c?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4b4c68374d7dd1f3edc86cf1b8f5e53c?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Medium","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Medium","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/medium","url":"https://daily.dev/sources/medium_js"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/tape-ml-detection-classification-and-now-segmentation-of-the-sota-level-in-0-1m-parameters-4e2gachee","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"machine-learning,deep-learning,computer-vision,object-detection","timeRequired":"PT11M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Medium","item":"https://daily.dev/sources/medium_js"},{"@type":"ListItem","position":3,"name":"TAPe+ML: detection, classification, and now segmentation of the SOTA‑level in 0.1M parameters"}]} ```