# Grasp Detection on Cornell Grasping Dataset (Image-wise split) Source: [Grasp Detection on Cornell Grasping Dataset (Image-wise split)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:40:45.818Z Evaluation of Grasp Detection performance on the Cornell Grasping Dataset using the image-wise split, measured by Point and Rectangle 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 | Grasp Detection | | Dataset (including any stated split) | Cornell Grasping Dataset (Image-wise split) | | Metric | Detection Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 25 of 25 | | 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 - [Detection Accuracy](): Higher is better; unit: PERCENT. - [Point Metric](): Higher is better; unit: PERCENT. - [Rectangle Metric](): Higher is better; unit: PERCENT. ## Recorded results | Method | Detection Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Multi-Modal Grasp Predictor | 89.21 | Not recorded | 68fa63801e209131ce9fef79 | | Uni-modal Grasp Predictor | 88.84 | Not recorded | 68fa63801e209131ce9fef79 | | Uni-modal Grasp Predictor | 88.53 | Not recorded | 68fa63801e209131ce9fef79 | | STEM-CaRFs (Selective Grasp) | 88.2 | Not recorded | 68fa63801e209131ce9fef79 | | MultiGrasp Detection | 88 | Not recorded | 68fa61b71e209131ce9f5a89 | | AlexNet, MultiGrasp | 88 | Not recorded | 68fa63801e209131ce9fef79 | | Multi-Modal Grasp Predictor | 86.44 | Not recorded | 68fa63801e209131ce9fef79 | | Regression + Classification | 85.5 | Not recorded | 68fa61b71e209131ce9f5a89 | | Two-stage closed-loop, with penalty | 85.3 | Not recorded | 68fa63801e209131ce9fef79 | | Uni-modal Grasp Predictor | 84.76 | Not recorded | 68fa63801e209131ce9fef79 | | Direct Regression | 84.4 | Not recorded | 68fa61b71e209131ce9f5a89 | | Lenz et al. | 73.9 | Not recorded | 68fa61b71e209131ce9f5a89 | | SAE, struct. reg. two-stage | 73.9 | Not recorded | 68fa63801e209131ce9fef79 | | Jiang et al. | 60.5 | Not recorded | 68fa61b71e209131ce9f5a89 | | Fast Search | 60.5 | Not recorded | 68fa63801e209131ce9fef79 | | Chance | 6.7 | Not recorded | 68fa61b71e209131ce9f5a89 | | Chance | 6.7 | Not recorded | 68fa63801e209131ce9fef79 | ### Multi-Modal Grasp Predictor Detection Accuracy: 89.21. Result ID: 698c743537894ef7d9ce7d07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50x2","Activation":"ReLU, ReLU, ReLU","Input":"RGB-D"}. This record does not assert independent reproduction. ### Uni-modal Grasp Predictor Detection Accuracy: 88.84. Result ID: 698c743537894ef7d9ce7d04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Activation":"ReLU, ReLU","Input":"RGB"}. This record does not assert independent reproduction. ### Uni-modal Grasp Predictor Detection Accuracy: 88.53. Result ID: 698c743537894ef7d9ce7d05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Activation":"tanh, ReLU","Input":"RGD"}. This record does not assert independent reproduction. ### STEM-CaRFs (Selective Grasp) Detection Accuracy: 88.2. Result ID: 698c743537894ef7d9ce7d02. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MultiGrasp Detection Detection Accuracy: 88. Result ID: 698c665437894ef7d9ce1dc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Time / image":null}. This record does not assert independent reproduction. ### AlexNet, MultiGrasp Detection Accuracy: 88. Result ID: 698c743537894ef7d9ce7d00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Multi-Modal Grasp Predictor Detection Accuracy: 86.44. Result ID: 698c743537894ef7d9ce7d06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50x2","Classifier":"linear SVM","Input":"RGB-D"}. This record does not assert independent reproduction. ### Regression + Classification Detection Accuracy: 85.5. Result ID: 698c665437894ef7d9ce1dc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Time / image":"76 ms"}. This record does not assert independent reproduction. ### Two-stage closed-loop, with penalty Detection Accuracy: 85.3. Result ID: 698c743537894ef7d9ce7d01. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Uni-modal Grasp Predictor Detection Accuracy: 84.76. Result ID: 698c743537894ef7d9ce7d03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Classifier":"SVM","Input":"RGB","Note":"Baseline"}. This record does not assert independent reproduction. ### Direct Regression Detection Accuracy: 84.4. Result ID: 698c665437894ef7d9ce1dc5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Time / image":null}. This record does not assert independent reproduction. ### Lenz et al. Detection Accuracy: 73.9. Result ID: 698c665437894ef7d9ce1dc4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Time / image":"13.5 sec"}. This record does not assert independent reproduction. ### SAE, struct. reg. two-stage Detection Accuracy: 73.9. Result ID: 698c743537894ef7d9ce7cff. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Jiang et al. Detection Accuracy: 60.5. Result ID: 698c665437894ef7d9ce1dc3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Time / image":null}. This record does not assert independent reproduction. ### Fast Search Detection Accuracy: 60.5. Result ID: 698c743537894ef7d9ce7cfe. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Chance Detection Accuracy: 6.7. Result ID: 698c665437894ef7d9ce1dc2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Time / image":null}. This record does not assert independent reproduction. ### Chance Detection Accuracy: 6.7. Result ID: 698c743537894ef7d9ce7cfd. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.