Nevin Baiju’s Post

Thought about inverse Kinematic engine? well here is one , made in C++ by me , V1 , uses IK, FK , Complex Analysis and Vector Analysis, base lock feature ,(SDL2 is used for rendering, should I implement it in OpenGL/vulkan? , well , that's maybe even more complex for me , Maybe I should make a snake game using inverse kinematics, it would be fun, wouldn't it? repo link: https://lnkd.in/dyrTA7iz #sdl2 #sdl2_gfx #gamedev #simulation #inversekinematics #forwardkinematics #complexanalysis #vectoranalysis #physicsengine #kinematicsengine

🎧 Unbox the VG710-M: An ASMR Experience! Ever wondered what cutting-edge technology sounds like? In this ASMR unboxing video, we unveil the VG710-M, your all-in-one connectivity solution for public transport systems. Experience the satisfying clicks of M12 connectors, the smooth unwrap of precision packaging, and the subtle beeps of innovation—all while exploring the features that make the VG710-M a game-changer: 🚍 Stable M12 interface connections 📡 Advanced GNSS for precise positioning 🛠️ Integrated vehicle diagnostics for optimized fleet management 💻 Custom development capabilities (Python, C/C++, Docker) 🌐 Remote management via DeviceLive 💡 Sit back, relax, and discover how the VG710-M is revolutionizing public transport connectivity. 🎥 Watch now! #ASMRUnboxing #ASMR #VG710M #PublicTransport #5G #Connectivity #ITS #ITxPT #Innovation #InHandNetworks

Day 19 of GFG160: Solved "Minimum Characters to Add for Palindrome" today! This challenge focused on determining the minimum characters required to make a string a palindrome. By leveraging the KMP Algorithm, I efficiently identified the longest palindromic suffix, transforming the problem into an elegant O(n) solution. Every challenge is an opportunity to refine my algorithmic thinking, and this one was no exception. It highlighted the value of pre-processing techniques in optimizing solutions. Excited to see what tomorrow holds as the journey continues! 🚀 #GFG160 #Day19 #geekstreak2024 #DSA #CodingChallenge #ProblemSolving

New tests with #LTXV, installed on #ComfyUI and run with #RTX4090. The goal of the tests was to increase the #duration of the individual #clips while maintaining good #temporal c#oherence, with image-to-video generations. Here is an interesting trick learned on #Reddit: feed the cropped image with the exact resolution to obtain and apply a small blur (1px) at the beginning of the workflow. Apparently, since the model is trained on moving images with natural motion blur, this trick significantly helps the generation of the clips.

Here is the video for implementation of gates using multiplexer. #verilog #vlsi #digitalelectronics #rtl_design link: https://lnkd.in/gpisuBZi

Real-Time Object Detection: The "Enemy" Within! Just wrapped up a FUN project on real-time object detection using C++ (YOLOv8 & ONNX)—but plot twist: the model thinks everything is an ENEMY! 😂 My client trained the model, and let’s just say... if it sees a chair, Enemy. A cat? Enemy. A banana? Enemy. 🚨 Basically, my job was to test this masterpiece in C++ while questioning all my life choices. But hey, at least it runs fast! And in the end, isn’t that all that matters? 😎 Daniel EtukudoIrum Zahra AwanMuhammad Rizwan Munawar Felix Sam Nanor Ultralytics #C++ #YOLOv8 #ONNX #ObjectDetection #EverythingIsEnemy

Did you know that there are SIMD instructions/intrinsics that are used more than others? Did you know that this also applies to pairs and triplets of SIMD instructions? Did you also know that there are >800 AVX512 intrinsics and >400 Arm Neon intrinsics that are essentially unused and probably just waste space in the silicon? Depending on your POV, these might sound reasonable or absurd assumptions, but until now we were not aware of such a research carried out against large codebases to make any such claim. We at VectorCamp did that, across a range of 13k open source Git repositories on Github (obviously we could not access the closed source ones!) and scanned the SIMD intrinsics usage for all major SIMD engines: SSE4.2, AVX2, AVX512 for x86, Arm Neon and Power VSX. Other engines will follow. Over a period of a few weeks, we collected large amounts of data for all these SIMD engines and did some statistics on these data. Some of these results we present as a blog: https://lnkd.in/dFQj2mq7 Kudos to our Giorgos T. for his excellent work on this project. Expect more from us on SIMD! #vectorization #simd

After a while trying to improve how to control shows from stage, multiprotocol remotes are ready to go to the next step. Succesfully tested on MadMapper, TouchDesigner, QLab, Resolume and Chamsys MagicQ. Protocols: MIDI, OSC, BLE, ArtNet, sACN, Serial.

Free virtual manipulatives with Polypad! Check out this featured #edTech tool at TechKnowMath.com!

Learn more about interpolation of regressors in Monolix and Simulx, including linear interpolation, which is new in MonolixSuite 2024. Watch the feature of the week video on linear interpolation here: https://lnkd.in/ecPQwzxZ