Startseite Neuigkeiten Neueste Forschungsarbeiten Tutorials Datensätze Wiki SOTA LLM-Modelle GPU-Rangliste Veranstaltungen

Deutsch

Referring Expression Segmentation On Davis

Metriken

Ju0026F 1st frame

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	Ju0026F 1st frame	Paper Title	Repository
DEVA (ReferFormer)	66.3	Tracking Anything with Decoupled Video Segmentation
UNINEXT-H	72.5	Universal Instance Perception as Object Discovery and Retrieval
URVOS + Refer-Youtube-VOS + ft. DAVIS	51.63	URVOS: Unified Referring Video Object Segmentation Network with a Large-Scale Benchmark
SgMg	63.3	Spectrum-guided Multi-granularity Referring Video Object Segmentation
RefVOS	44.5	RefVOS: A Closer Look at Referring Expressions for Video Object Segmentation
Khoreva et al.	39.3	Video Object Segmentation with Language Referring Expressions	-
VATEX	-	Vision-Aware Text Features in Referring Image Segmentation: From Object Understanding to Context Understanding
HTR	65.6	Temporally Consistent Referring Video Object Segmentation with Hybrid Memory
HINet	50.2	Hierarchical interaction network for video object segmentation from referring expressions	-
RefVOS	45.1	SynthRef: Generation of Synthetic Referring Expressions for Object Segmentation
ReferFormer	61.1	Language as Queries for Referring Video Object Segmentation
URVOS + Refer-Youtube-VOS	46.85	URVOS: Unified Referring Video Object Segmentation Network with a Large-Scale Benchmark
UniVS(Swin-L)	59.4?	UniVS: Unified and Universal Video Segmentation with Prompts as Queries
URVOS	44.1	URVOS: Unified Referring Video Object Segmentation Network with a Large-Scale Benchmark
PolyFormer-B	60.9	PolyFormer: Referring Image Segmentation as Sequential Polygon Generation
HyperSeg	71.2	HyperSeg: Towards Universal Visual Segmentation with Large Language Model
SafaRi-B	61.3	SafaRi:Adaptive Sequence Transformer for Weakly Supervised Referring Expression Segmentation	-
RefVOS + SynthRef-YouTube-VIS	45.3	SynthRef: Generation of Synthetic Referring Expressions for Object Segmentation

0 of 18 row(s) selected.