📊 概要
20件のポスト（リプライ除く）を4カテゴリに分類しました。
| カテゴリ | 件数 | 割合 |
|---|---|---|
| 🏗️ 3D再構成・SLAM | 11 | ██████████ 55% |
| 🚗 自動運転 | 5 | █████ 25% |
| 🤖 ロボティクス | 1 | █ 5% |
| 📄 論文紹介 | 3 | ███ 15% |
🏆 人気トップ3
🥇 1位
| RT | 35 |
| Like | 145 |
| Views | 11000 |
Loc-NeRF is a real-time robot localization method combining Monte Carlo localization and NeRF. It uses a pre-trained NeRF model as a map and an RGB camera, offering faster localization without requiring an initial pose estimate.
🔗 ポストを見る
🥈 2位
| RT | 23 |
| Like | 100 |
| Views | 7000 |
A new LiDAR Simulation Library! Code of “Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library” (ICRA 2023)
🔗 ポストを見る
🥉 3位
| RT | 13 |
| Like | 86 |
| Views | 6200 |
This paper introduces a new Eikonal formulation for continuous 3D scene completion in LiDAR point clouds, offering robustness to hyperparameters and the possibility for semantic Eikonal scene completion with minor modifications. [ICRA 2023] LODE
🔗 ポストを見る
📂 カテゴリ別ハイライト
🏗️ 3D再構成・SLAM
11件のポスト
- Loc-NeRF is a real-time robot localization method combining Monte Carlo localization and NeRF. It uses a pre-trained NeRF model as a map and an RGB ca… (♥ 145)
- This paper introduces a new Eikonal formulation for continuous 3D scene completion in LiDAR point clouds, offering robustness to hyperparameters and t… (♥ 86)
- Addressing SLAM in unknown environments involves estimating trajectory, landmark, and data associations. By dividing the problem into inner (trajector… (♥ 65)
- The paper introduces riSAM, a method for online incremental SLAM using Graduated Non-Convexity, achieving online efficiency and outperforming current … (♥ 57)
- The paper presents a novel SLAM strategy combining RGB-D and LiDAR sensors, improving performance in public benchmarks without data association or env… (♥ 39)
🚗 自動運転
5件のポスト
- A new LiDAR Simulation Library! Code of “Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library” (ICRA 2023) (♥ 100)
- This paper revisits the eight-point algorithm for two-view geometry and introduces a self-supervised deep convolutional neural network for improved no… (♥ 61)
- A novel RRT*-inspired online informative path planning algorithm is introduced to improve robot autonomy by addressing local minima and global coverag… (♥ 50)
- The paper presents a LiDAR-based semantic map method for autonomous driving, using a BEV pyramid decoder and Camera-to-LiDAR distillation, enhancing p… (♥ 49)
- The proposed approach uses sequential LiDAR scans to create augmented objects for self-supervised pre-training. Fine-tuning on downstream tasks reduce… (♥ 18)
🤖 ロボティクス
1件のポスト
📄 論文紹介
3件のポスト
- DiffMimic uses differentiable physics simulators for efficient motion mimicking, achieving faster, stable convergence and outperforming RL-based metho… (♥ 83)
- TrafficGen is a data-driven method that learns from real-world driving data to generate realistic traffic scenarios. [ICRA 2023] “TrafficGen: Learning… (♥ 67)
- The DeRaining network, Sparse Transformer (DRSformer), enhances image deraining by adaptively selecting self-attention values and using a mixed-scale … (♥ 10)