Papers with Code 아카이브
PointGoal Navigation
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 1
- 보존된 평가 기록
- 2
Point goal navigation is a task where an agent navigates to a specific location, defined by coordinates (x, y), in an environment using sensory inputs like vision or depth, without a pre-built map, relying on real-time perception and decision-making.
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
MPVO: Motion-Prior based Visual Odometry for PointGoal Navigation
MOPA: Modular Object Navigation with PointGoal Agents
Emergence of Maps in the Memories of Blind Navigation Agents
Comparison of Model-Free and Model-Based Learning-Informed Planning for PointGoal Navigation
VER: Scaling On-Policy RL Leads to the Emergence of Navigation in Embodied Rearrangement
Unsupervised Visual Odometry and Action Integration for PointGoal Navigation in Indoor Environment
Is Mapping Necessary for Realistic PointGoal Navigation?
Embodied Navigation at the Art Gallery
Image-based Navigation in Real-World Environments via Multiple Mid-level Representations: Fusion Models, Benchmark and Efficient Evaluation
Benchmarking Augmentation Methods for Learning Robust Navigation Agents: the Winning Entry of the 2021 iGibson Challenge
Realistic PointGoal Navigation via Auxiliary Losses and Information Bottleneck
Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI
The Surprising Effectiveness of Visual Odometry Techniques for Embodied PointGoal Navigation
Out of the Box: Embodied Navigation in the Real World
GridToPix: Training Embodied Agents with Minimal Supervision
Large Batch Simulation for Deep Reinforcement Learning
How to Train PointGoal Navigation Agents on a (Sample and Compute) Budget
Auxiliary Tasks Speed Up Learning PointGoal Navigation
Learning to Explore using Active Neural SLAM
Learning to Move with Affordance Maps
Sim2Real Predictivity: Does Evaluation in Simulation Predict Real-World Performance?
DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames
Habitat: A Platform for Embodied AI Research
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.