RemoteCLIPOptions
Options & Configuration · AiDotNet.VisionLanguage.Encoders
Configuration options for the RemoteCLIP model.
Inheritance: ModelOptions → ContrastiveEncoderOptions → RemoteCLIPOptions
For Beginners
RemoteCLIP is CLIP trained specifically to understand satellite and aerial images. It knows what “urban area”, “agricultural land”, “forest”, “water body” etc. look like from above, and can classify and search through satellite imagery using text queries.
How It Works
RemoteCLIP (Liu et al., 2023) adapts CLIP for remote sensing imagery. It is fine-tuned on curated remote sensing image-text datasets covering satellite imagery, aerial photography, and geospatial data. It supports tasks like zero-shot scene classification, image-text retrieval, and object counting in remote sensing contexts.
Constructors
| Constructor | Summary |
|---|---|
RemoteCLIPOptions() |
Initializes default RemoteCLIP options. |
RemoteCLIPOptions(RemoteCLIPOptions other) |
Initializes a new instance by copying from another instance. |
Properties
| Property | Summary |
|---|---|
ContrastiveLossType LossType { get; set; } |
Gets or sets the contrastive loss type. |
DomainSpecialization Domain { get; set; } |
Gets or sets the domain specialization. |
bool UseMultiScale { get; set; } |
Gets or sets whether to use multi-scale features for remote sensing. |
double GroundSampleDistance { get; set; } |
Gets or sets the ground sample distance in meters per pixel. |
string PromptTemplate { get; set; } |
Gets or sets the remote sensing specific text prompt template. |