Hacker News with Generative AI: Computer Vision

Dr. TVAM – Inverse Rendering for Tomographic Volumetric Additive Manufacturing (github.com/rgl-epfl)
Dr.TVAM is an inverse rendering framework for tomographic volumetric additive manufacturing.
3D Printing, Computer Vision, Artificial Intelligence, Software, Research
43 points by roflmaostc 9 hours ago | 8 comments
Issues with color spaces and perceptual brightness (johnaustin.io)
Unlike RGB, the CIELab color spaceAnd the more modern variants like CIECAM02 and Oklab. is designed to be perceptually uniform.
Color Spaces, Perceptual Brightness, Computer Vision
111 points by todsacerdoti 19 hours ago | 63 comments
Show HN: I created a PoC for live descriptions of the surroundings for the blind (github.com/o40)
I wanted to see if I could create a low-cost tool for the blind to get live description of the scene in front of a camera.
Accessibility, Computer Vision, Open Source, Blindness
73 points by o40 14 days ago | 25 comments
Auto Smiley – Computer vision smile generator (2010) (fffff.at)
Auto Smiley is a computer vision application that runs in the background while you work. The software analyzes your face while you are working and if it detects a smile it sends the the ascii smiley face letters “: )” as keyboard presses to the front most application. Auto Smiley has many uses from just straight up convenience to enforcing honesty in your online communication :)
Computer Vision, Software, Productivity, Humor
10 points by bo0tzz 15 days ago | 0 comments
Show HN: DeepFace – A lightweight deep face recognition library for Python (github.com/serengil)
DeepFace is a lightweight face recognition and facial attribute analysis (age, gender, emotion and race) framework for python.
Python, Deep Learning, Face Recognition, Computer Vision
257 points by serengil 15 days ago | 46 comments
DeepSeek-VL2: MoE Vision-Language Models for Advanced Multimodal Understanding (github.com/deepseek-ai)
Introducing DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL.
Computer Vision, Generative AI, Artificial Intelligence
36 points by selvan 17 days ago | 7 comments
Tenstorrent Wormhole Series (corsix.org)
A company called Tenstorrent design and sell PCIe cards for AI acceleration. At the time of writing, they've recently started shipping their Wormhole n150s and Wormhole n300s cards.
Hardware, AI, Computer Vision, Deep Learning, Machine Learning
9 points by FL33TW00D 17 days ago | 0 comments
Reflecting on o3 "beating ARC": are we reliving the ImageNet 2012 moment again? (ycombinator.com)
AlexNet came and blown everything out of the water. Then you can reflect how much [a lot] progress there has been since 2012 till now just on this little dataset.<p>o3 beating ARC is such a harder dataset, I don't even want to compare them. So how much progress there will be from just this?<p>Next 10 years gonna be bonkers.
Artificial Intelligence, Machine Learning, Computer Vision, Benchmarking, Progress
7 points by artificialprint 18 days ago | 3 comments
All You Need Is 4x 4090 GPUs to Train Your Own Model (sabareesh.com)
My journey into Large Language Models (LLMs) began with the excitement of seeing ChatGPT in action. I started by exploring diffusion models, drawn to their ability to create beautiful visuals. However, working on an M1 chip had its limitations, which motivated me to build a custom rig with an NVIDIA 4090 GPU. As I continued to explore LLMs and experimented with multi-agent systems, I came to realize the importance of mastering the fundamentals.
Artificial Intelligence, Hardware, Training, Computer Vision
125 points by sabareesh 20 days ago | 120 comments
Armada: Augmented Reality for Robot Manipulation and Robot-Free Data Acquisition (arxiv.org)
Teleoperation for robot imitation learning is bottlenecked by hardware availability. Can high-quality robot data be collected without a physical robot?
Robotics, Artificial Intelligence, Computer Vision, Data Acquisition
6 points by sandwichsphinx 21 days ago | 0 comments
Reverse Video Search (mixpeek.com)
Reverse video search allows us to use a video clip as an input for a query against videos that have been indexed in a vector store.
Search Engines, Computer Vision
24 points by Beefin 21 days ago | 10 comments
Language Model as Visual Explainer (arxiv.org)
In this paper, we present Language Model as Visual Explainer LVX, a systematic approach for interpreting the internal workings of vision models using a tree-structured linguistic explanation, without the need for model training.
Generative AI, Machine Learning, Computer Vision
5 points by PaulHoule 21 days ago | 0 comments
DeepSeek-v3 Technical Report [pdf] (github.com/deepseek-ai)
Deep Learning, Computer Vision, Research Papers
26 points by pongogogo 22 days ago | 2 comments
Trying out QvQ – Qwen's new visual reasoning model (simonwillison.net)
I thought we were done for major model releases in 2024, but apparently not: Alibaba’s Qwen team just dropped the Apache 2.0 licensed QvQ-72B-Preview, “an experimental research model focusing on enhancing visual reasoning capabilities”.
Generative AI, Artificial Intelligence, Computer Vision
236 points by simonw 24 days ago | 82 comments
INFP: Audio-Driven Interactive Head Generation in Dyadic Conversations (grisoon.github.io)
We present INFP, an audio-driven interactive head generation framework for dyadic conversations. Given the dual-track audio in dyadic conversations and a single portrait image of arbitrary agent, our framework can dynamically synthesize verbal, non-verbal and interactive agent videos with lifelike facial expressions and rhythmic head pose movements. Additionally, our framework is lightweight yet powerful, making it practical in instant communication scenarios such as the video conferencing. INFP denotes our method is Interactive, Natural, Flash and Person-generic.
Generative AI, Computer Vision, Video Conferencing, Audio Processing
29 points by nnx 28 days ago | 21 comments
FastVideo: a lightweight framework for accelerating large video diffusion models (github.com/hao-ai-lab)
FastVideo is a lightweight framework for accelerating large video diffusion models.
Deep Learning, Video Processing, Computer Vision, Generative AI, Software
110 points by zhisbug 31 days ago | 24 comments
Fine-tuning a vision model to recognize break dance power moves (bawolf.com)
I’ve been looking for a way to combine software engineering with my break practice and stumbled upon the idea of fine-tuning a vision model on power moves.
Computer Vision, Machine Learning, Dance, Breakdancing
7 points by bryantwolf 31 days ago | 3 comments
Nvidia Jetson Orin Nano Super [video] (youtube.com)
Nvidia, Hardware, Embedded Systems, Video, Computer Vision
65 points by ralusek 31 days ago | 51 comments
Meta's new Video Understanding Multimodal Model used Qwen model for training (arxiv.org)
Despite the rapid integration of video perception capabilities into Large Multimodal Models (LMMs), the underlying mechanisms driving their video understanding remain poorly understood.
Meta, Artificial Intelligence, Video Processing, Computer Vision
7 points by BUFU 32 days ago | 1 comments
Veo 2: Our video generation model (deepmind.google)
Veo creates videos with realistic motion and high quality output, up to 4K. Explore different styles and find your own with extensive camera controls.
Video Generation, Artificial Intelligence, Deep Learning, Google, Computer Vision
587 points by mvoodarla 32 days ago | 327 comments
Representing Long Volumetric Video with Temporal Gaussian Hierarchy (zju3dv.github.io)
This paper aims to address the challenge of reconstructing long volumetric videos from multi-view RGB videos.
Computer Vision, Video Processing, 3D Reconstruction
7 points by pr337h4m 34 days ago | 0 comments
Llama.cpp Now Supports Qwen2-VL (Vision Language Model) (github.com/ggerganov)
This PR implements the Qwen2VL model as requested at #9246 . The main changes include:
Open Source, Computer Vision
155 points by BUFU 34 days ago | 50 comments
OpenAI announces Advanced Voice with Vision [video] (youtube.com)
OpenAI, Artificial Intelligence, Voice Recognition, Computer Vision
13 points by BoorishBears 36 days ago | 1 comments
AI pioneer Fei-Fei Li has a vision for computer vision (ieee.org)
AI pioneer Fei-Fei Li says to unlock visual intelligence, we need to respect the fact that "the world is 3D."
Computer Vision, Artificial Intelligence, 3D Vision
75 points by samizdis 36 days ago | 55 comments
Attribute Extraction from Images Using DSPy (langtrace.ai)
DSPy recently added support for VLMs in beta. A quick thread on attributes extraction from images using DSPy. For this example, we will see how to extract useful attributes from screenshots of websites
Image Processing, Machine Learning, Computer Vision, Artificial Intelligence, Software
9 points by kakaly0403 39 days ago | 0 comments
Long Convolutions via Polynomial Multiplication (hazyresearch.stanford.edu)
We’ve been writing a series of papers (1, 2, 3) that have at their core so-called long convolutions, with an aim towards enabling longer-context models. These are different from the 3x3 convolutions people grew up with in vision because well, they are longer–in some cases, with filters as long as the whole sequence. A frequent line of questions we get is about what these long convolutions are and how we compute them efficiently, so we put together a short tutorial.
Machine Learning, Computer Vision
46 points by cochlear 40 days ago | 13 comments
Convolutional Neural Network Visualization [video] (youtube.com)
Machine Learning, Computer Vision, Deep Learning, Video
6 points by simonjgreen 42 days ago | 0 comments
Show HN: Real-Time YOLO Object Detection in Elixir: Fast, Simple, Extensible (github.com/poeticoding)
YOLO is an Elixir library designed to simplify object detection by providing seamless integration of YOLO models. With this library, you can efficiently utilize the power of YOLO for real-time object detection.
Elixir, Object Detection, Computer Vision, Open Source, Show HN
96 points by alvises 42 days ago | 11 comments
Moondream 0.5B: The Smallest Vision-Language Model (moondream.ai)
Today, we’re thrilled to announce the launch of Moondream 0.5B, the latest addition to our family of open-source AI models. With only 0.5 billion parameters, Moondream 0.5B is the world's smallest Vision-Language Model (VLM), designed to unlock AI's potential on edge devices and mobile platforms. It builds on the success of its predecessor, Moondream 2B, and is also released under the flexible Apache License, ensuring accessibility for everyone.
Artificial Intelligence, Open Source, Computer Vision
14 points by BUFU 43 days ago | 3 comments
PaliGemma 2: Powerful Vision-Language Models, Simple Fine-Tuning (googleblog.com)
Building custom, advanced AI that can "see" used to be a complex and resource-intensive endeavor. Not anymore. This past May, we launched PaliGemma, the first vision-language model in the Gemma family, taking a significant step toward making class-leading visual AI more accessible. Now, we're thrilled to introduce PaliGemma 2, the next evolution in tunable vision-language models.
Generative AI, Computer Vision, Machine Learning, Language Models
218 points by meetpateltech 43 days ago | 27 comments