Change baseline_m to baseline_metric

Co-authored-by: Sarat Poluri <sarat.chandra.poluri@intel.com>

Author: saratpoluri

mapping/src/vggt_model.py

@@ -469,7 +469,7 @@ def _runModelInference(self, images_tensor: torch.Tensor) -> Dict[str, Any]:
 
     return predictions
 
-  def _baseline_m_from_camera_locations(self, camera_locations, camera_ids=None) -> float:
+  def _baseline_metric_from_camera_locations(self, camera_locations, camera_ids=None) -> float:
     """
     Robustly compute a metric baseline (meters) from camera_locations.
 
@@ -588,16 +588,16 @@ def _processOutputs(self, predictions: Dict[str, Any], original_sizes: List[tupl
       camera_to_world_list.append(c2w)
 
     # --- SCALE FIX: compute metric baseline from provided camera_locations ---
-    baseline_m = self._baseline_m_from_camera_locations(camera_locations, camera_ids=camera_ids)
+    baseline_m = self._baseline_metric_from_camera_locations(camera_locations, camera_ids=camera_ids)
 
-    if baseline_m <= 0:
+    if baseline_metric <= 0:
       log.warning("VGGT: camera_locations missing/invalid; skipping metric scaling (scale will be arbitrary).")
 
-    if baseline_m > 0 and len(camera_to_world_list) >= 2:
+    if baseline_metric > 0 and len(camera_to_world_list) >= 2:
       b_units = self._baseline_units(camera_to_world_list[0], camera_to_world_list[1])
       if b_units > 1e-6:
-        s = baseline_m / b_units
-        log.info(f"Scaling VGGT outputs by s={s:.6f} (baseline {baseline_m:.6f}m / {b_units:.6f} units)")
+        s = baseline_metric / b_units
+        log.info(f"Scaling VGGT outputs by s={s:.6f} (baseline {baseline_metric:.6f}m / {b_units:.6f} units)")
 
         # scale camera translations
         for k in range(len(camera_to_world_list)):