https://github.com/microsoft/UniTAB
공식 코드UniTAB: Unifying Text and Box Outputs for Grounded Vision-Language Modeling
Papers with Code 아카이브
Visual Grounding
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 3
- 평가표
- 4
- 보존된 평가 기록
- 20
Visual Grounding (VG) aims to locate the most relevant object or region in an image, based on a natural language query. The query can be a phrase, a sentence, or even a multi-round dialogue. There are three main challenges in VG:
- What is the main focus in a query?
- How to understand an image?
- How to locate an object?
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/microsoft/unilm/tree/master/kosmos-2
공식 코드Kosmos-2: Grounding Multimodal Large Language Models to the World
https://github.com/mightyzau/infmllm
공식 코드https://github.com/minghangz/resvg
공식 코드ResVG: Enhancing Relation and Semantic Understanding in Multiple Instances for Visual Grounding
https://github.com/mismatchquest/mismatchquest
공식 코드Mismatch Quest: Visual and Textual Feedback for Image-Text Misalignment
https://github.com/moured/refchartqa
공식 코드RefChartQA: Grounding Visual Answer on Chart Images through Instruction Tuning
https://github.com/mr-bigworth/mmca
공식 코드https://github.com/mr-neko/nice
공식 코드NICE: Improving Panoptic Narrative Detection and Segmentation with Cascading Collaborative Learning
https://github.com/mrlearnedtoad/scaneru
공식 코드ScanERU: Interactive 3D Visual Grounding based on Embodied Reference Understanding
https://github.com/mshukor/vicha
공식 코드Efficient Vision-Language Pretraining with Visual Concepts and Hierarchical Alignment
https://github.com/next-chatv/next-chat
공식 코드https://github.com/nini0919/diffpng
공식 코드Exploring Phrase-Level Grounding with Text-to-Image Diffusion Model
https://github.com/niveditarufus/CMSVG
공식 코드Cosine meets Softmax: A tough-to-beat baseline for visual grounding
https://github.com/ofa-sys/ofa
공식 코드https://github.com/ojedaf/merlim
공식 코드Behind the Magic, MERLIM: Multi-modal Evaluation Benchmark for Large Image-Language Models
https://github.com/olivesgatech/introspective-learning
공식 코드Introspective Learning : A Two-Stage Approach for Inference in Neural Networks
https://github.com/om-ai-lab/groundvlp
공식 코드https://github.com/om-ai-lab/vlm-r1
공식 코드VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model
https://github.com/onmyoji-xiao/3dvg_secg
공식 코드SeCG: Semantic-Enhanced 3D Visual Grounding via Cross-modal Graph Attention
https://github.com/onmyoji-xiao/AS3D
공식 코드https://github.com/opendatalab/h2rsvlm
공식 코드VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis
https://github.com/opendatalab/vhm
공식 코드VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis
https://github.com/opendrivelab/drivelm
공식 코드https://github.com/opengvlab/internvl
공식 코드https://github.com/openrobotlab/ease
공식 코드Evolving Symbolic 3D Visual Grounder with Weakly Supervised Reflection
https://github.com/openrobotlab/embodiedscan
공식 코드MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations
https://github.com/openrobotlab/vlm-grounder
공식 코드https://github.com/openspaceai/uvltrack
공식 코드Unifying Visual and Vision-Language Tracking via Contrastive Learning
https://github.com/orionbench/orionbench
공식 코드OrionBench: A Benchmark for Chart and Human-Recognizable Object Detection in Infographics
https://github.com/petrichor625/talk2car_cavg
공식 코드https://github.com/pku-rl/copl
공식 코드Visual Grounding for Object-Level Generalization in Reinforcement Learning
https://github.com/ppjmchen/ham
공식 코드Learning Point-Language Hierarchical Alignment for 3D Visual Grounding
https://github.com/qizekun/ShapeLLM
공식 코드ShapeLLM: Universal 3D Object Understanding for Embodied Interaction
https://github.com/qumengxue/siri-vg
공식 코드SiRi: A Simple Selective Retraining Mechanism for Transformer-based Visual Grounding
https://github.com/qwenlm/qwen-vl
공식 코드Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
https://github.com/qzp2018/MCLN
공식 코드Multi-branch Collaborative Learning Network for 3D Visual Grounding
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.