# Sub-activity Labeling on CAD-120 Source: [Sub-activity Labeling on CAD-120]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:52:45.397Z Evaluation of Sub-activity Labeling performance on the CAD-120 dataset, reporting micro and macro precision and recall metrics. 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 | Sub-activity Labeling | | Dataset (including any stated split) | CAD-120 | | Metric | Micro Precision/Recall | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Micro Precision/Recall](): Higher is better; unit: PERCENT. - [Macro Precision](): Higher is better; unit: PERCENT. - [Macro Recall](): Higher is better; unit: PERCENT. ## Recorded results | Method | Micro Precision/Recall | Result date | Report paper | | --- | --- | --- | --- | | full model | 86 | Not recorded | 68fa60f41e209131ce9f1f03 | | no object interactions | 85.3 | Not recorded | 68fa60f41e209131ce9f1f03 | | full model with tracking | 82.5 | Not recorded | 68fa60f41e209131ce9f1f03 | | no temporal interactions | 76 | Not recorded | 68fa60f41e209131ce9f1f03 | | sub-activity only | 71.9 | Not recorded | 68fa60f41e209131ce9f1f03 | | full, multi-seg learning | 68.2 | Not recorded | 68fa60f41e209131ce9f1f03 | | full, 1 segment. (best) | 66.6 | Not recorded | 68fa60f41e209131ce9f1f03 | | full, 1 segment. (averaged) | 64.3 | Not recorded | 68fa60f41e209131ce9f1f03 | | full, multi-seg learning + tracking | 63.4 | Not recorded | 68fa60f41e209131ce9f1f03 | | SVM multiclass | 58 | Not recorded | 68fa60f41e209131ce9f1f03 | | image only | 56.2 | Not recorded | 68fa60f41e209131ce9f1f03 | | max class | 29.2 | Not recorded | 68fa60f41e209131ce9f1f03 | ### full model Micro Precision/Recall: 86. Result ID: 698bcb3c7a967eddb77167b9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### no object interactions Micro Precision/Recall: 85.3. Result ID: 698bcb3c7a967eddb77167b8. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### full model with tracking Micro Precision/Recall: 82.5. Result ID: 698bcb3c7a967eddb77167ba. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### no temporal interactions Micro Precision/Recall: 76. Result ID: 698bcb3c7a967eddb77167b7. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### sub-activity only Micro Precision/Recall: 71.9. Result ID: 698bcb3c7a967eddb77167b6. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### full, multi-seg learning Micro Precision/Recall: 68.2. Result ID: 698bcb3c7a967eddb77167bd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"None (Estimated)"}. This record does not assert independent reproduction. ### full, 1 segment. (best) Micro Precision/Recall: 66.6. Result ID: 698bcb3c7a967eddb77167bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"None (Estimated)"}. This record does not assert independent reproduction. ### full, 1 segment. (averaged) Micro Precision/Recall: 64.3. Result ID: 698bcb3c7a967eddb77167bc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"None (Estimated)"}. This record does not assert independent reproduction. ### full, multi-seg learning + tracking Micro Precision/Recall: 63.4. Result ID: 698bcb3c7a967eddb77167be. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Temporal segmentation":"None (Estimated)"}. This record does not assert independent reproduction. ### SVM multiclass Micro Precision/Recall: 58. Result ID: 698bcb3c7a967eddb77167b5. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### image only Micro Precision/Recall: 56.2. Result ID: 698bcb3c7a967eddb77167b4. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. ### max class Micro Precision/Recall: 29.2. Result ID: 698bcb3c7a967eddb77167b3. [Reporting paper]() Recorded evaluation configuration: {"Temporal segmentation":"Ground-truth"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.