# Novel Class Discovery on SemanticPOSS Split 2 Source: [Novel Class Discovery on SemanticPOSS Split 2]() Collection index: [llms.txt]() Data updated: 2026-06-08T07:46:10.894Z Evaluation of Novel Class Discovery performance on Split 2 of the SemanticPOSS dataset, reporting metrics for novel classes, known classes, and overall results across various object categories. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Novel Class Discovery | | Dataset (including any stated split) | SemanticPOSS Split 2 | | Metric | Bike Performance | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Bike Performance](): Higher is better; unit: PERCENT. - [Building Performance](): Higher is better; unit: PERCENT. - [Car Performance](): Higher is better; unit: PERCENT. - [Cone Performance](): Higher is better; unit: PERCENT. - [Fence Performance](): Higher is better; unit: PERCENT. - [Ground Performance](): Higher is better; unit: PERCENT. - [Person Performance](): Higher is better; unit: PERCENT. - [Plants Performance](): Higher is better; unit: PERCENT. - [Pole Performance](): Higher is better; unit: PERCENT. - [Rider Performance](): Higher is better; unit: PERCENT. - [Traffic Sign Performance](): Higher is better; unit: PERCENT. - [Trash Can Performance](): Higher is better; unit: PERCENT. - [Trunk Performance](): Higher is better; unit: PERCENT. - [Novel Class Performance (Avg)](): Higher is better; unit: PERCENT. - [Known Class Performance (Avg)](): Higher is better; unit: PERCENT. - [Overall Performance](): Higher is better; unit: PERCENT. ## Recorded results | Method | Bike Performance | Result date | Report paper | | --- | --- | --- | --- | | HyperNCD | 45.5 | Not recorded | 6a264ca690966ade919beae2 | | DASL | 45.3 | Not recorded | 6a264ca690966ade919beae2 | | EUMS | 40.1 | Not recorded | 6a264ca690966ade919beae2 | | SNOPS | 38.4 | Not recorded | 6a264ca690966ade919beae2 | | NOPS | 37.2 | Not recorded | 6a264ca690966ade919beae2 | ### HyperNCD Bike Performance: 45.5. Result ID: 6a26733ffbc373592c22f5a3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"MinkowskiUNet-34C","Learning Protocol":"Novel Class Discovery"}. This record does not assert independent reproduction. ### DASL Bike Performance: 45.3. Result ID: 6a26733ffbc373592c22f5a2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"MinkowskiUNet-34C","Learning Protocol":"Novel Class Discovery"}. This record does not assert independent reproduction. ### EUMS Bike Performance: 40.1. Result ID: 6a26733ffbc373592c22f59f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"MinkowskiUNet-34C","Learning Protocol":"Novel Class Discovery"}. This record does not assert independent reproduction. ### SNOPS Bike Performance: 38.4. Result ID: 6a26733ffbc373592c22f5a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"MinkowskiUNet-34C","Learning Protocol":"Novel Class Discovery"}. This record does not assert independent reproduction. ### NOPS Bike Performance: 37.2. Result ID: 6a26733ffbc373592c22f5a0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"MinkowskiUNet-34C","Learning Protocol":"Novel Class Discovery"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.