Myteam commited on
Commit
b730f2f
Β·
verified Β·
1 Parent(s): 4e66c8d

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +47 -88
app.py CHANGED
@@ -4,9 +4,8 @@ Abnormal-Behaviour Detector (VideoMAE)
4
 
5
  Tabs
6
  ────
7
- 1. **Video file** – upload / record β†’ annotated video, static plot, beep.
8
- 2. **Live stream** – webcam feed + rolling blue/red bar‐plot + numeric prob.
9
- Sensitivity slider now goes from 0.0 β†’ 1.0, default 0.0.
10
  """
11
 
12
  import os, cv2, tempfile, numpy as np, matplotlib
@@ -26,12 +25,12 @@ processor = AutoImageProcessor.from_pretrained(MODEL_DIR, use_fast=False)
26
  model.eval()
27
 
28
  # ── Constants ─────────────────────────────────────────────────────────
29
- CLIP = 16 # frames per inference clip
30
- STRIDE = 1 # evaluate on every new frame
31
- STREAM_FPS = 15 # polling ~15Γ—/sec
32
- HIST_LEN = 90 # keep last 90 prob entries
33
 
34
- # ── Helpers for β€œVideo file” tab ────────────────────────────────────────
35
  def sliding_window(path, clip_len=CLIP, stride=STRIDE):
36
  cap, frames = cv2.VideoCapture(path), []
37
  fps = cap.get(cv2.CAP_PROP_FPS) or 25
@@ -44,40 +43,35 @@ def sliding_window(path, clip_len=CLIP, stride=STRIDE):
44
 
45
  probs = []
46
  for s in range(0, len(frames) - clip_len + 1, stride):
47
- clip = [cv2.resize(frames[i], (224, 224))
48
- for i in range(s, s + clip_len)]
49
- inp = processor(clip, return_tensors="pt").to(DEVICE)
50
  with torch.no_grad():
51
- p = torch.softmax(model(**inp).logits, dim=1)[0,1].item()
52
- probs.append(p)
53
  return frames, probs, fps
54
 
 
55
  def make_outputs(frames, probs, fps, thresh):
56
- # annotate
57
  h, w = frames[0].shape[:2]
58
  for i,p in enumerate(probs):
59
  if p > thresh:
60
  for f in range(i, min(i+CLIP, len(frames))):
61
  cv2.rectangle(frames[f], (0,0),(w-1,h-1),(255,0,0),6)
62
 
63
- # save video
64
- vid = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False).name
65
- vw = cv2.VideoWriter(vid, cv2.VideoWriter_fourcc(*"mp4v"), fps, (w,h))
66
  for f in frames:
67
  vw.write(cv2.cvtColor(f, cv2.COLOR_RGB2BGR))
68
  vw.release()
69
 
70
- # static plot
71
  t = np.arange(len(probs)) / fps
72
  fig, ax = plt.subplots(figsize=(5,2))
73
  colors = ["red" if p>thresh else "blue" for p in probs]
74
  ax.bar(t, probs, color=colors, width=1/fps)
75
  ax.set_ylim(0,1); ax.set_xlabel("sec"); ax.set_ylabel("fight prob")
76
  fig.tight_layout()
77
- plot = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
78
- fig.savefig(plot, dpi=120); plt.close(fig)
79
 
80
- # beep
81
  alarm = None
82
  if any(p>thresh for p in probs):
83
  sr = 16000
@@ -85,7 +79,8 @@ def make_outputs(frames, probs, fps, thresh):
85
  alarm = tempfile.NamedTemporaryFile(suffix=".wav",delete=False).name
86
  sf.write(alarm, 0.2*np.sin(2*np.pi*440*t_beep), sr)
87
 
88
- return vid, plot, alarm
 
89
 
90
  def run_pipeline(file_path, thresh):
91
  if not file_path or not os.path.exists(file_path):
@@ -93,91 +88,55 @@ def run_pipeline(file_path, thresh):
93
  frames, probs, fps = sliding_window(file_path)
94
  return make_outputs(frames, probs, fps, thresh)
95
 
96
-
97
- # ── Helpers for β€œLive stream” tab ─────────────────────────────────────
98
- def plot_history(history, thresh):
99
  fig, ax = plt.subplots(figsize=(5,2))
100
- colors = ["red" if p>thresh else "blue" for p in history]
101
- ax.bar(np.arange(len(history)), history, color=colors)
102
  ax.set_ylim(0,1); ax.set_xlabel("recent clips"); ax.set_ylabel("fight prob")
103
- fig.tight_layout()
104
- return fig
105
 
106
  def live_handler(frame, state, history, thresh):
107
  state.append(frame)
108
  if len(state) < CLIP:
109
  return state, frame, history, plot_history(history, thresh), 0.0
110
 
111
- # infer every frame
112
  clip = [cv2.resize(f, (224,224)) for f in list(state)]
113
  inp = processor(clip, return_tensors="pt").to(DEVICE)
114
- with torch.no_grad():
115
- p = torch.softmax(model(**inp).logits, dim=1)[0,1].item()
116
-
117
  history.append(p)
118
- if len(history) > HIST_LEN:
119
- history.pop(0)
120
-
121
- # annotate & overlay
122
- disp = frame.copy()
123
- h, w = disp.shape[:2]
124
- if p > thresh:
125
- cv2.rectangle(disp, (0,0),(w-1,h-1),(255,0,0),6)
126
- cv2.putText(disp, f"fight prob: {p:.2f}",
127
- (10,h-15), cv2.FONT_HERSHEY_SIMPLEX, 0.7,
128
- (255,255,255), 2, cv2.LINE_AA)
129
 
130
- return state, disp, history, plot_history(history, thresh), p
 
 
 
131
 
 
132
 
133
- # ── Build Gradio UI ────────────────────────────────────────────────────
134
  with gr.Blocks(title="Abnormal Behaviour Detection") as demo:
 
 
135
 
136
- # Sensitivity slider now 0.0 β†’ 1.0, default 0.0
137
- thresh_slider = gr.Slider(
138
- minimum=0.0, maximum=1.0, step=0.01, value=0.0,
139
- label="Sensitivity threshold (lower β†’ more sensitive)"
140
- )
141
-
142
- # ─ Tab 1: Video file ─────────────────────────────────────────────
143
  with gr.Tab("Video file"):
144
- gr.Markdown(
145
- "Upload a video (.mp4/.avi/.mov) or record one. "
146
- "Adjust **Sensitivity**, then click **Analyze**."
147
- )
148
- file_in = gr.Video(label="Upload / record", format="mp4",
149
- webcam_options=gr.WebcamOptions())
150
- vid_out = gr.Video(label="Annotated video")
151
- plot_out = gr.Image(label="Fight probability curve")
152
- beep_out = gr.Audio(label="Alarm")
153
-
154
- gr.Button("Analyze").click(
155
- fn=run_pipeline,
156
- inputs=[file_in, thresh_slider],
157
- outputs=[vid_out, plot_out, beep_out],
158
- )
159
-
160
- # ─ Tab 2: Live stream ────────────────────────────────────────────
161
- with gr.Tab("Live stream"):
162
- gr.Markdown(
163
- "Live webcam detection. Border & bar‐plot turn red whenever "
164
- "`prob > Sensitivity`. See the **Current prob** below in real time."
165
- )
166
-
167
- cam_in = gr.Image(sources="webcam", streaming=True,
168
- type="numpy", label="Webcam")
169
- cam_out = gr.Image(streaming=True, label="Live annotated")
170
- plot_live = gr.Plot(label="Fight probability (rolling)")
171
- num_prob = gr.Number(label="Current prob", precision=2)
172
 
 
 
 
 
173
  frame_state = gr.State(value=deque(maxlen=CLIP))
174
  prob_state = gr.State(value=[])
175
-
176
- cam_in.stream(
177
- fn=live_handler,
178
- inputs=[cam_in, frame_state, prob_state, thresh_slider],
179
- outputs=[frame_state, cam_out, prob_state, plot_live, num_prob],
180
- stream_every=1/STREAM_FPS,
181
- )
182
 
183
  demo.queue().launch()
 
4
 
5
  Tabs
6
  ────
7
+ 1. Video file – upload / record β†’ annotated video, static plot, beep.
8
+ 2. Live stream – webcam feed + rolling bar‑plot + numeric prob; sensitivity slider defaults to 0.0.
 
9
  """
10
 
11
  import os, cv2, tempfile, numpy as np, matplotlib
 
25
  model.eval()
26
 
27
  # ── Constants ─────────────────────────────────────────────────────────
28
+ CLIP = 16 # frames per clip
29
+ STRIDE = 1 # evaluate each frame
30
+ STREAM_FPS = 15 # ~15 updates/sec
31
+ HIST_LEN = 90 # history length for rolling plot
32
 
33
+ # ── Video-file pipeline (unchanged) ──────────────────────────────────
34
  def sliding_window(path, clip_len=CLIP, stride=STRIDE):
35
  cap, frames = cv2.VideoCapture(path), []
36
  fps = cap.get(cv2.CAP_PROP_FPS) or 25
 
43
 
44
  probs = []
45
  for s in range(0, len(frames) - clip_len + 1, stride):
46
+ clip = [cv2.resize(frames[i], (224,224)) for i in range(s, s+clip_len)]
47
+ inp = processor(clip, return_tensors="pt").to(DEVICE)
 
48
  with torch.no_grad():
49
+ probs.append(torch.softmax(model(**inp).logits, dim=1)[0,1].item())
 
50
  return frames, probs, fps
51
 
52
+
53
  def make_outputs(frames, probs, fps, thresh):
 
54
  h, w = frames[0].shape[:2]
55
  for i,p in enumerate(probs):
56
  if p > thresh:
57
  for f in range(i, min(i+CLIP, len(frames))):
58
  cv2.rectangle(frames[f], (0,0),(w-1,h-1),(255,0,0),6)
59
 
60
+ vid_out = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False).name
61
+ vw = cv2.VideoWriter(vid_out, cv2.VideoWriter_fourcc(*"mp4v"), fps, (w,h))
 
62
  for f in frames:
63
  vw.write(cv2.cvtColor(f, cv2.COLOR_RGB2BGR))
64
  vw.release()
65
 
 
66
  t = np.arange(len(probs)) / fps
67
  fig, ax = plt.subplots(figsize=(5,2))
68
  colors = ["red" if p>thresh else "blue" for p in probs]
69
  ax.bar(t, probs, color=colors, width=1/fps)
70
  ax.set_ylim(0,1); ax.set_xlabel("sec"); ax.set_ylabel("fight prob")
71
  fig.tight_layout()
72
+ plot_out = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
73
+ fig.savefig(plot_out, dpi=120); plt.close(fig)
74
 
 
75
  alarm = None
76
  if any(p>thresh for p in probs):
77
  sr = 16000
 
79
  alarm = tempfile.NamedTemporaryFile(suffix=".wav",delete=False).name
80
  sf.write(alarm, 0.2*np.sin(2*np.pi*440*t_beep), sr)
81
 
82
+ return vid_out, plot_out, alarm
83
+
84
 
85
  def run_pipeline(file_path, thresh):
86
  if not file_path or not os.path.exists(file_path):
 
88
  frames, probs, fps = sliding_window(file_path)
89
  return make_outputs(frames, probs, fps, thresh)
90
 
91
+ # ── Live-stream helpers ───────────────────────────────────────────────
92
+ def plot_history(hist, thresh):
 
93
  fig, ax = plt.subplots(figsize=(5,2))
94
+ colors = ["red" if p>thresh else "blue" for p in hist]
95
+ ax.bar(np.arange(len(hist)), hist, color=colors)
96
  ax.set_ylim(0,1); ax.set_xlabel("recent clips"); ax.set_ylabel("fight prob")
97
+ fig.tight_layout(); return fig
98
+
99
 
100
  def live_handler(frame, state, history, thresh):
101
  state.append(frame)
102
  if len(state) < CLIP:
103
  return state, frame, history, plot_history(history, thresh), 0.0
104
 
 
105
  clip = [cv2.resize(f, (224,224)) for f in list(state)]
106
  inp = processor(clip, return_tensors="pt").to(DEVICE)
107
+ with torch.no_grad(): p = torch.softmax(model(**inp).logits, dim=1)[0,1].item()
 
 
108
  history.append(p)
109
+ if len(history) > HIST_LEN: history.pop(0)
 
 
 
 
 
 
 
 
 
 
110
 
111
+ disp = frame.copy(); h, w = disp.shape[:2]
112
+ if p > thresh: cv2.rectangle(disp, (0,0),(w-1,h-1),(255,0,0),6)
113
+ cv2.putText(disp, f"fight prob: {p:.2f}", (10,h-15), cv2.FONT_HERSHEY_SIMPLEX,
114
+ 0.7, (255,255,255),2,cv2.LINE_AA)
115
 
116
+ return state, disp, history, plot_history(history, thresh), p
117
 
118
+ # ── UI setup ──────────────────────────────────────────────────────────
119
  with gr.Blocks(title="Abnormal Behaviour Detection") as demo:
120
+ thresh_slider = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, value=0.0,
121
+ label="Sensitivity threshold (lower β†’ more sensitive)")
122
 
 
 
 
 
 
 
 
123
  with gr.Tab("Video file"):
124
+ gr.Markdown("Upload a video or record one, adjust Sensitivity, then Analyze.")
125
+ file_in = gr.Video(format="mp4", webcam_options=gr.WebcamOptions())
126
+ vid_out = gr.Video(); plot_out = gr.Image(); beep_out = gr.Audio()
127
+ gr.Button("Analyze").click(run_pipeline,
128
+ inputs=[file_in, thresh_slider],
129
+ outputs=[vid_out, plot_out, beep_out])
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
130
 
131
+ with gr.Tab("Live stream"):
132
+ gr.Markdown("Live webcam detection. Border+plot turn red when p>threshold.")
133
+ cam_in = gr.Image(sources="webcam", streaming=True, type="numpy")
134
+ cam_out = gr.Image(streaming=True); plot_live = gr.Plot(); num_p = gr.Number()
135
  frame_state = gr.State(value=deque(maxlen=CLIP))
136
  prob_state = gr.State(value=[])
137
+ cam_in.stream(live_handler,
138
+ inputs=[cam_in, frame_state, prob_state, thresh_slider],
139
+ outputs=[frame_state, cam_out, prob_state, plot_live, num_p],
140
+ stream_every=1/STREAM_FPS)
 
 
 
141
 
142
  demo.queue().launch()