👆iSeg: Interactive 3D Segmentation👆

👉 iSeg: interactive segmentation technique for 3D shapes operating entirely in 3D. It accepts both positive/negative clicks directly on the shape's surface, indicating inclusion & exclusion of regions.

👉Review https://t.ly/tyFnD
👉Paper https://lnkd.in/dydAz8zp
👉Project https://lnkd.in/de-h6SRi
👉Code (coming)

👗 Neural Bodies with Clothes 👗

👉Neural-ABC is a novel parametric model based on neural implicit functions that can represent clothed human bodies with disentangled latent spaces for ID, clothing, shape, and pose.

👉Review https://t.ly/Un1wc
👉Project https://lnkd.in/dhDG6FF5
👉Paper https://lnkd.in/dhcfK7jZ
👉Code https://lnkd.in/dQvXWysP

🔌 BodyMAP: human body & pressure 🔌

👉#Nvidia (+CMU) unveils BodyMAP, the new SOTA in predicting body mesh (3D pose & shape) and 3D applied pressure on the human body. Source Code released, Dataset coming 💙

👉Review https://t.ly/8926S
👉Project bodymap3d.github.io/
👉Paper https://lnkd.in/gCxH4ev3
👉Code https://lnkd.in/gaifdy3q

🧞 XComposer2: 4K Vision-Language 🧞

👉InternLMXComposer2-4KHD brings LVLM resolution capabilities up to 4K HD (3840×1600) and beyond. Authors: Shanghai AI Lab, CUHK, SenseTime & Tsinghua. Source Code & Models released 💙

👉Review https://t.ly/GCHsz
👉Paper arxiv.org/pdf/2404.06512.pdf
👉Code github.com/InternLM/InternLM-XComposer

⚛️ Flying w/ Photons: Neural Render ⚛️

👉Novel neural rendering technique that seeks to synthesize videos of light propagating through a scene from novel, moving camera viewpoints. Pico-Seconds time resolution!

👉Review https://t.ly/ZqL3a
👉Paper arxiv.org/pdf/2404.06493.pdf
👉Project anaghmalik.com/FlyingWithPhotons/
👉Code github.com/anaghmalik/FlyingWithPhotons

☄️ Tracking Any 2D Pixels in 3D ☄️

👉 SpatialTracker lifts 2D pixels to 3D using monocular depth, represents the 3D content of each frame efficiently using a triplane representation, and performs iterative updates using a transformer to estimate 3D trajectories.

👉Review https://t.ly/B28Cj
👉Paper https://lnkd.in/d8ers_nm
👉Project https://lnkd.in/deHjtZuE
👉Code https://lnkd.in/dMe3TvFT

🪐YOLO-CIANNA: Neural Astro🪐

👉 CIANNA is a general-purpose deep learning framework for (but not only for) astronomical data analysis. Source Code released 💙

👉Review https://t.ly/441XS
👉Paper arxiv.org/pdf/2402.05925.pdf
👉Code github.com/Deyht/CIANNA
👉Wiki github.com/Deyht/CIANNA/wiki

🧤Neuro MusculoSkeletal-MANO🧤

👉SJTU unveils MusculoSkeletal-MANO, novel musculoskeletal system with a learnable parametric hand model. Source Code announced 💙

👉Review https://t.ly/HOQrn
👉Paper arxiv.org/pdf/2404.10227.pdf
👉Project https://ms-mano.robotflow.ai/
👉Code announced (no repo yet)

⚽SoccerNET: Athlete Tracking⚽

👉SoccerNet Challenge is a novel high level computer vision task that is specific to sports analytics. It aims at recognizing the state of a sport game, i.e., identifying and localizing all sports individuals (players, referees, ..) on the field.

👉Review https://t.ly/Mdu9s
👉Paper arxiv.org/pdf/2404.11335.pdf
👉Code github.com/SoccerNet/sn-gamestate

🎲 Articulated Objs from MonoClips 🎲

👉REACTO is the new SOTA to address the challenge of reconstructing general articulated 3D objects from single monocular video

👉Review https://t.ly/REuM8
👉Paper https://lnkd.in/d6PWagij
👉Project https://lnkd.in/dpg3x4tm
👉Repo https://lnkd.in/dRZWj6_N

🪼 All You Need is SAM (+Flow) 🪼

👉Oxford unveils the new SOTA for moving object segmentation via SAM + Optical Flow. Two novel models & Source Code announced 💙

👉Review https://t.ly/ZRYtp
👉Paper https://lnkd.in/d4XqkEGF
👉Project https://lnkd.in/dHpmx3FF
👉Repo coming: https://github.com/Jyxarthur/

🛞 6Img-to-3D driving scenarios 🛞

👉EPFL (+ Continental) unveils 6Img-to-3D, novel transformer-based encoder-renderer method to create 3D onbounded outdoor driving scenarios with only six pics

👉Review https://shorturl.at/dZ018
👉Paper arxiv.org/pdf/2404.12378.pdf
👉Project 6img-to-3d.github.io/
👉Code github.com/continental/6Img-to-3D

🌹 Physics-Based 3D Video-Gen 🌹

👉PhysDreamer, a physics-based approach that leverages the object dynamics priors learned by video generation models. It enables realistic 3D interaction with objects

👉Review https://t.ly/zxXf9
👉Paper arxiv.org/pdf/2404.13026.pdf
👉Project physdreamer.github.io/
👉Code github.com/a1600012888/PhysDreamer

🎡 NER-Net: Seeing at Night-Time 🎡

👉Huazhong (+Beijing) unveils a novel event-based nighttime imaging solution under non-uniform illumination, plus a paired multi-illumination level real-world dataset. Repo online, code coming 💙

👉Review https://t.ly/Z9JMJ
👉Paper arxiv.org/pdf/2404.11884.pdf
👉Repo github.com/Liu-haoyue/NER-Net
👉Clip https://www.youtube.com/watch?v=zpfTLCF1Kw4

🌊 FlowMap: dense depth video 🌊

👉MIT (+CSAIL) unveils FlowMap, a novel E2E differentiable method that solves for precise camera poses, camera intrinsics, and perframe dense depth of a video sequence. Source Code released 💙

👉Review https://t.ly/CBH48
👉Paper arxiv.org/pdf/2404.15259.pdf
👉Project cameronosmith.github.io/flowmap
👉Code github.com/dcharatan/flowmap

👗TELA: Text to 3D Clothed Human👗

👉 TELA is a novel approach for the new task of clothing disentangled 3D human model generation from texts. This novel approach unleashes the potential of many downstream applications (e.g., virtual try-on).

👉Review https://t.ly/6N7JV
👉Paper https://arxiv.org/pdf/2404.16748
👉Project https://jtdong.com/tela_layer/
👉Code https://github.com/DongJT1996/TELA

🪷 Tunnel Try-on: SOTA VTON 🪷

👉"Tunnel Try-on", the first diffusion-based video virtual try-on model that demonstrates SOTA performance in complex scenarios. No code announced :(

👉Review https://t.ly/joMtJ
👉Paper arxiv.org/pdf/2404.17571
👉Project mengtingchen.github.io/tunnel-try-on-page/

🏝️1000x Scalable Neural 3D Fields🏝️

👉Highly-scalable neural 3D Fields: 1000x reductions in memory maintaining speed/quality: 10 MB vs. 10 GB! Code released 💙

👉Review https://t.ly/sLTK5
👉Paper https://lnkd.in/dEYM8-t2
👉Project https://lnkd.in/djptdujx
👉Code https://lnkd.in/dcCnFZ2n

🌐3D Scenes w/ Depth Inpainting🌐

👉Oxford announced two novel contributions to the field of 3D scene generation: a new benchmark and a novel depth completion model. 🤗-Demo and Source Code released💙

👉Review https://t.ly/BKiny
👉Paper arxiv.org/pdf/2404.19758
👉Project research.paulengstler.com/invisible-stitch/
👉Code github.com/paulengstler/invisible-stitch
👉Demo huggingface.co/spaces/paulengstler/invisible-stitch

🌊 Diffusive 3D Human Recovery 🌊

👉The Rutgers University unveils ScoreHMR at #CVPR24; novel approach for 3D human pose and shape reconstruction. Impressive results.

👉Review https://t.ly/G0k2D
👉Paper https://arxiv.org/pdf/2403.09623
👉Code https://github.com/statho/ScoreHMR
👉Project https://statho.github.io/ScoreHMR/

🏷️DiffMOT (#CVPR24): diffusion-MOT🏷️

👉DiffMOT is a novel real-time diffusion-based MOT approach to tackle the complex nonlinear motion. Impressive results & Source Code released💙

👉Review https://t.ly/ztlHi
👉Paper https://lnkd.in/d4K3c-nt
👉Project https://diffmot.github.io/
👉Code github.com/Kroery/DiffMOT