Papers with Code 아카이브
Depth Estimation
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 10
- 평가표
- 14
- 보존된 평가 기록
- 58
Depth Estimation is the task of measuring the distance of each pixel relative to the camera. Depth is extracted from either monocular (single) or stereo (multiple views of a scene) images. Traditional methods use multi-view geometry to find the relationship between the images. Newer methods can directly estimate depth by minimizing the regression loss, or by learning to generate a novel view from a sequence. The most popular benchmarks are KITTI and NYUv2. Models are typically evaluated according to a RMS metric.
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
Scaling 4D Representations
Flowing from Words to Pixels: A Framework for Cross-Modality Evolution
Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusion
Prompting Depth Anything for 4K Resolution Accurate Metric Depth Estimation
Foundation Models Meet Low-Cost Sensors: Test-Time Adaptation for Rescaling Disparity for Zero-Shot Metric Depth Estimation
PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts
Depth-Centric Dehazing and Depth-Estimation from Real-World Hazy Driving Video
V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations
FSFM: A Generalizable Face Security Foundation Model via Self-Supervised Facial Representation Learning
ViPOcc: Leveraging Visual Priors from Vision Foundation Models for Single-View 3D Occupancy Prediction
MAL: Cluster-Masked and Multi-Task Pretraining for Enhanced xLSTM Vision Performance
Cross-View Completion Models are Zero-shot Correspondence Estimators
T-SVG: Text-Driven Stereoscopic Video Generation
Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos
Dense Depth from Event Focal Stack
Utilizing Multi-step Loss for Single Image Reflection Removal
BLADE: Single-view Body Mesh Learning through Accurate Depth Estimation
Balancing Shared and Task-Specific Representations: A Hybrid Approach to Depth-Aware Video Panoptic Segmentation
Event fields: Capturing light fields at high speed, resolution, and dynamic range
Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View Scene Reconstruction
On-Device Self-Supervised Learning of Low-Latency Monocular Depth from Only Events
Driv3R: Learning Dense 4D Reconstruction for Autonomous Driving
SphereUFormer: A U-Shaped Transformer for Spherical 360 Perception
GVDepth: Zero-Shot Monocular Depth Estimation for Ground Vehicles based on Probabilistic Cue Fusion
TACO: Learning Multi-modal Action Models with Synthetic Chains-of-Thought-and-Action
PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion
SimC3D: A Simple Contrastive 3D Pretraining Framework Using RGB Images
MT3DNet: Multi-Task learning Network for 3D Surgical Scene Reconstruction
MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videos
DualPM: Dual Posed-Canonical Point Maps for 3D Shape and Pose Reconstruction
LAA-Net: A Physical-prior-knowledge Based Network for Robust Nighttime Depth Estimation
Align3R: Aligned Monocular Depth Estimation for Dynamic Videos
MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction
Dense Scene Reconstruction from Light-Field Images Affected by Rolling Shutter
Perception Tokens Enhance Visual Reasoning in Multimodal Language Models
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.